Abstract: Petroleum-Based Hydrocarbon Investigations: Are Groundwater Resources Adequately Protected by Present Practice?

Philip B. Chandler

Conventional demonstration of petroleum-based hydrocarbon soil contamination through soil matrix sampling has severe systemic limitations long ignored. Aromatic volatile organic compound (VOC) soil contamination is frequently understated by so-called "standard practice", which is to obtain a relative few soil matrix samples in sometimes dubious fashion, often from excavations left open to rainfall accumulation, analyze with low-priced TPH methodology for petroleum-based hydrocarbons and project the spatial distribution of all hydrocarbon and related compound contamination. It has been shown that this may not necessarily be sufficient to evaluate threat to ground water. Intensive regulatory investigations at the San Gabriel and San Fernando Superfund sites as well as at ot er locations in the Los Angeles area have produced a large sample set of soil gas measurements at a variety of source types, such as at underground waste oil and fuel tanks, aboveground waste and product tanks at refineries and used oil recyclers. These have yielded information on relatively mobile constituents, generally recognized as most threatening to drinking water resources once dissolved in ground water. Because these mobile constituents, mostly aromatic VOCs, such as benzene, toluene, ethylbenzene and xylenes readily partition into vapor phase representative analytical measurements could be made on soil gases, collected by relatively conservative methodologies, at a given site with little or no holding time losses. Moreover, gas phase halogenated VOCs were also frequently identif ed with respect to sites where only fuel or waste oils were to be expected. In many instances, collateral measurement of fixed and biogenic gases at the same time provided a proxy measure of heavier hydrocarbon distribution. Multi-depth spatial patterns of gases more closely connected site discharge and source area than did sole reliance on soil matrix sampling and the vagaries of holding time losses over as much as a fourteen day period. This is a somewhat divergent application for soil gas from the oversold groundwater "plume-tracking" or the "reconnaissance-only" most commonly associated with it. As opposed to such conventional "wisdom", investigations were designed to evaluate waste discharge to the vadose zone, assess the lateral and vertical distribution of contamination and to det rmine if the ground water was being threatened by migration from surface and subsurface discharges. The data and information derived indicate that in order to folly protect groundwater resources, soil gases should be measured together with soil matrix samples at each site under investigation.