An official website of the United States government.

This is not the current EPA website. To navigate to the current EPA website, please go to www.epa.gov. This website is historical material reflecting the EPA website as it existed on January 19, 2021. This website is no longer updated and links to external websites and some internal pages may not work. More information »

# UST Technical Compendium: Release Detection

• Question 1: Please clarify EPA's UST regulations as they apply to a threshold value for declaring an UST or its piping to be leaking using a tightness test.
[November 11, 1988 letter from Jack Horner of Horner Creative Products, Inc.]
Answer: To be able to detect a 0.1 gallon per hour (gph) leak as required in the regulations at a statistically reliable level of confidence, the threshold must be smaller than 0.1 gph. The correct threshold depends on the particular leak detection system and UST system, but is traditionally and typically 0.05 gph.
[December 19, 1988 letter to Mr. Horner (PDF) (57 pp, 644 K)]
• Question 2: Is annual line testing required if a pressurized line is equipped with a permanent line monitoring device?
[June 1989 letter from Judith Spray of Pollulert Systems]
Answer: If an automatic line monitoring device meets the regulatory standard for a line tightness test, 0.1 gallon per hour at 1.5 times operating pressure, then it can be substituted for an annual line tightness test.  Note that here are also ways to comply that do not involve line tightness testing.
[June 29, 1989 letter to Ms. Spray (PDF) (57 pp, 644 K)]
• Question 3:  How do you convert a leak rate at one operating pressure to an equivalent leak rate at another pressure?
[February 1990 letter from Michael Bouton of Tracer Research Corporation]
Answer: The appropriate formula for the conversion is that the leak rate is proportional to the square root of the pressure drop ratio.
[February 28, 1990 letter to Mr. Bouton (PDF) (57 pp, 644 K)]
• Question 4: Do pressurized lines at UST sites that have monitoring wells around tank pits, but not along piping runs, also have to have an annual line pressure test?
[June 29, 1990 letter from Bill Birdwell of Tanknology Corporation International]
Answer: Pressurized lines must have automatic catastrophic leak detection backed-up by a monthly monitoring or annual line tightness test. If a tank excavation is intercepted by observation wells, but a pressurized line system extends beyond the designed reach of those monitoring wells, then an annual line tightness test or monthly monitoring is required.
[July 19, 1990 letter to Mr. Birdwell (PDF) (57 pp, 644 K)]
• Question 5: Can statistical inventory reconciliation (SIR) be used to comply with EPA's UST regulations, including requirements for UST-associated piping?
[April 10, 1991 letter from Deborah Talanian of Entropy Limited]
Answer: SIR is generally a "tank system" test. Thus, if properly performed for any particular site, a SIR method that demonstrated adequate performance under the EPA evaluation protocols may be an acceptable alternative to periodic line tightness testing. The line still must have a catastrophic line leak detector, however.
[May 10, 1991 letter to Ms. Talanian (PDF) (57 pp, 644 K)]
• Question 6: For purposes of EPA leak detection requirements what constitutes the portion of the tank that "routinely contains product"?
[June 1991 request from Region VIII]
Answer: EPA has determined it is protective of human health and the environment to be somewhat flexible about what portion of the upper part of the tank must be tested so that UST owners and operators can take full advantage of the different types of leak detection available. The "routinely contains product" language is intended to make clear that test methods do not have to test vent pipes, fill pipes, and fittings on top of the tank. Some simple rules of thumb about how far below these parts of the tank can be tested: (1) acoustic methods and SIR methods should not be a concern, as to the level tested; (2) ATG level-sensing methods should only be tested down the level at which the method was third-party evaluated; (3) major in-tank level monitoring service providers most often specify 85 to 95 percent full as their own protocol for testing and this is seen as meeting the routinely contains product provision; and (4) small business tanks with low product sales may test well below the top of the tank, if inventory data demonstrates restricted tank in filling practices that result in routinely low levels in the tank.
[July 25, 1991 memorandum to Region VIII (PDF) (57 pp, 644 K)]
[June 26, 1991 Attachment (PDF) (57 pp, 644 K)]
• Question 7: Can manual tank gauging be used as the sole method of leak detection for tanks larger than 550 gallons?
[January 22, 1990 from Priscilla Young of the American Petroleum Institute]
Answer: When conducted in accordance with the procedures described in the attachment, manual tank gauging meets the performance specifications under 40 CFR 280.43(h)(1) for tanks of nominal capacity of 1000 gallons or less  and, therefore, can be used as the sole means of leak detection.
[April 6, 1990 letter to Ms. Young (PDF) (57 pp, 644 K)]
• Question 8: Do the performance capabilities of "catastrophic" automatic line leak detectors have to be tested in the field on an annual basis?
[August 12, 1991 memo from Region VII]
Answer: As requirements for automatic line leak detectors (LLDs), LLDs must be tested annually in accordance with manufacturer's requirements but not necessarily tested to any particular leak rate. The standard of 3 gallons per hour at 10 pounds per square inch within 1 hour is not the standard for the annual test but rather for the initial performance evaluation.
[March 5, 1992 memorandum to Regional Program Managers (PDF) (57 pp, 644 K)]
• Question 9: Do systems incorporating flexible liners (bladders) and vacuum monitor systems meet the technical requirements?
[March 19, 1992 letter from John Hendershot of World Enviro Systems]
Answer: Flexible internally-fitted liner systems can be shown to meet the Federal requirements for release detection for both petroleum and hazardous substance USTs if certain conditions are met, including compatibility and automatic detection of a breach in either the outer tank or the inner liner. These systems cannot meet Federal requirements for upgrading or repairing existing UST systems.
[July 9, 1992 letter to Mr. Hendershot (PDF) (57 pp, 644 K)]
• Question 10: Please clarify whether the Federal underground storage tank regulations at 40 CFR 280.43 require inventory control with automatic tank gauges (ATGs).
[October 2, 1992 letter from Mr. Durgin of Veeder Root to David Ziegele (PDF) (57 pp, 644 K)]
Answer: Inventory control is not required, regardless of the installation date, if the ATG has been shown to meet the performance standard and the probabilities of detection and of false alarm.
[November 22, 1993 to Mr. Culp (PDF) (57 pp, 644 K)]
[April 18, 1989 Attachment (PDF) (57 pp, 644 K)]
• Question 11: If a facility is using statistical inventory reconciliation (SIR) monthly and a report is not conclusive, is the owner or operator out of compliance?
Answer: Yes, an UST system relying on monthly SIR with a report that does not conclusively indicate whether the system is leaking at 0.2 gallons per hour with a probability of detection of 0.95 and a probability of false alarm of 0.05 is not technically compliance with the release detection requirements. A lack of conclusive results may not be possible on a small percentage of tank data each month, for several reasons.  Because of this, EPA encourages states and regions to use discretion, and to consider the efforts of owners and operators to comply in assessing whether to enforce.
[November 18, 1993 memorandum to State and Regional Program Managers (PDF) (57 pp, 644 K)]
[February 7, 1994 letter to Mr. Hunt (PDF) (57 pp, 644 K)]
• Question 12: Request that EPA require siphon bars that join manifolded tank systems to be regulated with respect to leak detection and corrosion protection, and that inventory control not be an acceptable means of leak detection for a manifolded system.
[May 23, 1994 letter from Dale Tanke of the Office of the Illinois State Fire Marshal to Gerald Phillips, US EPA Region V.]
Answer: Siphon bars are already regulated in terms of leak detection and cathodic protection requirements as part of connected underground piping that routinely contains product. Due to the manner in which the syphons routinely operate, however, leak detection requirements are minimal, akin to that of safe suction lines.
As to leak detection with manifolded tanks, inventory control with periodic tightness testing is permissible until December 22, 1998 or until ten years subsequent to a new tank installation or upgrade.
[February 13, 1995 letter to Mr. Tanke (PDF) (57 pp, 644 K)]
• Question 13: In changing over from one form of leak detection to another, is it necessary to finish up a twelve-month cycle with the old method in order to be in compliance?
[October 4, 1994 letter from Robert Staab of the Circle K Corporation]
Answer: Changing from on acceptable leak detection method to another can be done at any time. It does not require the completion of a "cycle." However, it is important that all leak detection records are properly maintained in accordance with 40 CFR 280.45.
[February 7, 1995 letter to Mr. Staab (PDF) (57 pp, 644 K)]
• Question 14: Do I have to report a suspected release if my inventory control results from two consecutive months both exceed the performance standard, but one is short and the other is over?  Does this "confirm the initial result"?
[Sept. 18, 1995 letter from Robert Stabb, Circle K Stores Inc.]
Answer: Yes, reporting is required.  EPA interprets "confirm the initial result" at 280.50(c)(2) to mean a second exceedance, no matter whether the direction -- short or over -- is the same as the first month.
[December 12, 1995 letter to Mr. Stabb (PDF) (57 pp, 644 K)]
• Question 15: When can I use the combination of tightness testing every 5 years and inventory control for leak detection on my upgraded UST?
Answer: Owners/operators may begin using the combination of inventory control and 5-year tightness tests only after the entire UST system meets 1998 standards. This means that the UST system must have corrosion protection for both tank and piping, spill, and overfill protection. The combination of inventory control and tank tightness testing may be used until 10 years after the tank itself has met corrosion protection requirements or until 12/22/98, whichever is later. After this time, another monthly method for release detection must be used.
[July 25, 1997 memorandum to State and Regional Program Managers (PDF) (57 pp, 644 K)]

All references cited by the questions and answers above can be found together in the following resource: