ABSTRACT
The conference held at Argonne National Laboratory (Argonne) was a unique opportunity to share some ideas about geographical information systems (GIS) and predictive modeling in a relaxed and collegial, but extremely focused, setting. For me, it was an opportunity to express some thoughts I had about assessing the varying levels of confidence in predictive model results and tie those ideas to some innovative work that was ongoing at Argonne using adaptive sampling methods to characterize hazardouswaste sites. I was not exactly sure how the two ideas could be compatible, but the underlying concepts seemed worth investigating and sharing, if for no other reason than to obtain feedback from the many experts attending the conference. In writing this chapter, I continued my struggle to link the concepts proven effective in the context of hazardous-waste characterization and remediation with the practice of archaeological predictive modeling, particularly in a cultural resources-management context. Although I have not had the resources available to me to test the concepts in the field, I believe they are worth adding to the discussion on archaeological modeling brought forth in these proceedings.
