Skip to Main Content


The purpose of inversion is to determine the limits of prediction for model testing and risk assessment given the state of our knowledge (knowledge in the form of model assumptions, accuracy, and precision, and in the form of data distributions, types, accuracy, and precision). Inversion is a tool for making decisions. Should we do inversion? Are current models and assumptions sufficient to solve posed problems, or do we need to develop a new model? Are currently available data sufficient, or do we need to collect more or different data? And how do we decide?

The above plethora of queries arose during the committee discussions as general concerns, as did a large number of more specific questions. We record in Appendix 1 all of the questions that arose because the committee deliberations did not allow time for complete resolution of all issues. We recorded all of the questions in the hope that some, at least, will be answered in the future.

The question list of Appendix 1 is far too long to address in detail, so the committee focused on attempts to provide some pragmatic rules of operation to guide the actual technical use of inverse modeling. Although the rules set down in this paper are not a universal panacea, they do represent considerable practical experience with inverse models. As such, they are useful guides as one struggles to determine the strengths and weaknesses of inverse modeling.

You do not currently have access to this chapter.

Figures & Tables





Citing Books via

Close Modal
This Feature Is Available To Subscribers Only

Sign In or Create an Account

Close Modal
Close Modal