Effect of Monitoring Well Installations on Risk Assessments of Metals in Groundwater. B. S. Price, ASA Inc., P. O. Box 17532, Portland, ME 04101; and M. J. Carpenter, ABB Environmental Services, Inc., 110 Free Street, Portland, ME 04112
The quality of monitoring well installation often affects the concentrations of metals apparently in the groundwater and hence the outcome of the risk assessment. "Classical" groundwater sampling requires installation of wells with well slotting smaller than aquifer particle size, however, aquifer grain size is often not determined until after the well is installed. Most risks are calculated from unfiltered water samples, however, these are not necessarily representative of the aquifer, but are more dependent on the method of sampling. Data correlating turbidity or total suspended solids in samples with metal concentrations are used in an attempt to better characterize risks from metals in water. The metals that are most often affected are some of the more toxic metals, arsenic, lead and chromium, as well as measures of gross alpha and gross beta activities. Newer cost cutting methods for collecting groundwater samples often neglect the effects of sample collection on risk characterizations. The use of these "non-classical" groundwater sampling results requires careful planning to avoid characterizing risks that are not site-related but sampling related. Financial constraints often dictate the methods used to install wells, the type of well installed and the purging techniques. Balancing the cost of installing a $20,000 well against the possibility of an unneeded remediation system can discriminate against those with more limited funds. The pressures to implement field programs with "cost-cutting integrated sample collection" have resulted in unanticipated limitations on collected data that extend to their use in risk assessments and remedial design. A discussion supplemented by specific project examples will explore the question of representative risks and representative samples. Various techniques will also be described that can be used to obtain data more representative of the risks, regardless of the sample collection technique.