Abstract: Our study attempts to answer three basic questions:

1. From the available data, what can we say about the probability of a release occurring from an underground fuel storage tank? Thus far, we have been unable to find any study that reports on any population of USTs for which all the pertinent facts regarding leakage are known. Generally, tank closure activities are the dominant way in which leaks are discovered (Calif. SWRCB, 1997), thus making direct inference regarding leakage probability during normal operation unlikely. Analysis of national data (EPA Corrective Action Database 1996-1997)æincluding information such as number of active tanks, number of tanks closed, and number of confirmed releasesæallowed us to compute the number of releases per closure and per active tank across 66 reporting regions. Regression analysis of this data reveals that the upper bound for the probability of UST leakage is between 0.22 and 0.25.

2. Is there any reason to suspect that MTBE in reformulated fuels enhances the potential for leakage in UST systems? A comprehensive literature review of over 15 studies investigating compatibility of MTBE with UST and related (i.e. pipeline and automotive) systems indicates that no significant threat exists from the concentrations of MTBE likely to be present in reformulated fuels (Owens Corning, 1995; Alexander, Ferber, et al, 1994; Sun Refining, 1988). That is to say that, while neat MTBE may be incompatible with such systemsæspecifically with elastomeric sealing componentsæeffects seen at concentrations of up to 20% were not statistically different from those seen when MTBE concentration was 0%.

3. What effects can be expected on UST systems from MTBE alternatives? A preliminary review of literature pertaining to compatibility of ethanol, methanol, and other ether compounds such as ETBE and TAME has been made. Indications are that the alchohols tend to be more corrosive towards metals than pure hydrocarbon fuels and tend to swell some elastomers (API, 1990). There is also evidence of increased permeability through high-density polyethylene fuel tanks (SAE, 1992), as compared to pure gasoline. Review of this literature is as yet incomplete.