Sampling in the Smear Zone: Evaluation of Nondissolved Bias and Associated BTEX, MTBE, and TPH Concentrations in Ground Water Samples*

Authors


  • *

    This paper is a revised, peer-reviewed version of the material presented previously at the 2005 NGWA/API Petroleum Hydrocarbons Conference.

Abstract

At fuel release sites, source zone characterization and monitoring for benzene, toluene, ethylbenzene and xylenes, methyl-tert-butyl-ether, and total petroleum hydrocarbons as gasoline (TPHg) typically involve interpreting ground water data from grab (screening) samples and/or monitoring wells within the “smear zone,” where the soil pores contain both separate-phase petroleum and dissolved-phase constituents. Because the samples are collected within the smear zone, the ground water concentrations can be biased high by contributions from entrained nondissolved petroleum. The nondissolved bias is often not recognized by data users and can lead to erroneous interpretations and inputs for fate and transport modeling, mass flux calculations, risk assessments, etc., which can result in increased requirements for remediation or increased claims for damages in litigation. Ground water sample concentrations can be evaluated for potential bias by nondissolved petroleum using a phased approach that begins with comparison of the data to effective solubility concentrations for ethylbenzene, xylenes, and TPH derived from laboratory partitioning studies using fresh fuels. To reflect the reduction in effective solubility that results from weathering of the source, the ground water sample concentrations can also be compared to concentrations for samples that were collected with entrained sheen. In this data set of 47 gasoline release sites, the frequency of nondissolved bias in the reported ground water monitoring data ranged from 0.4% to 86%, depending on which concentration criterion was used as an indicator. If exceedances of the average concentration values for the laboratory partitioning and the sheen data are used, the frequency ranged from 9% to 20%. Also, reproducibility for duplicates collected from monitoring wells in the smear zone was unacceptable more than half of the time. This evaluation shows that a significant percentage of ground water monitoring data collected from the smear zone is unreliable for characterizing dissolved concentrations of petroleum constituents.

Ancillary