SAND96-0840                              Distribution
Unlimited Release                     Category UC-600
Printed April 1996
Adaptive Sampling Approach to Environmental Site
Characterization at Joliet Army Ammunition Plant:
Phase II Demonstration
Grace E. Bujewski
Environmental Restoration Technologies Department
Sandia National Laboratories
Albuquerque, NM 87185-0719
Robert L. Johnson
Argonne National Laboratory
Argonne, IL 60439
ABSTRACT
Adaptive sampling programs provide real opportunities to save considerable time and money when
characterizing hazardous waste sites. This Strategic Environmental Research and Development Program
(SERDP) project demonstrated two decision-support technologies, SitePlannerM and Plume, that can
facilitate the design and deployment of an adaptive sampling program. A demonstration took place at
Joliet Army Ammunition Plant (JAAP), and was unique in that it was tightly coupled with ongoing Army
characterization work at the facility, with close scrutiny by both state and federal regulators. The
demonstration was conducted in partnership with the Army Environmental Center's (AEC) Installation
Restoration Program and AEC's Technology Development Program. AEC supported researchers from
Tufts University who demonstrated innovative field analytical techniques for the analysis of TNT and
DNT.
SitePlannerm is an object-oriented database specifically designed for site characterization that
provides an effective way to compile, integrate, manage and display site characterization data as it is
being generated. Plume uses a combination of Bayesian analysis and geostatistics to provide technical
staff with the ability to quantitatively merge soft and hard information for an estimate of the extent of
contamination. Plume provides an estimate of contamination extent, measures the uncertainty
associated with the estimate, determines the value of additional sampling, and locates additional samples
so that their value is maximized.
The primary objectives identified for the demonstration were successfully accomplished. The
SERDP research team was able to generate graphics on the fly; to develop a conceptual model for the
site; to direct the sampling program; and to provide estimates of contamination for the JAAP remedial
action plan. We successfully coupled the adaptive sampling approach with an ongoing characterization
activity that had close regulatory oversight. We also identified key technical or logistical issues. The
JAAP demonstration showed that the adaptive sampling approach, as well as the specific technological
decision-support tools, are acceptable within the current regulatory framework.