OPR-PPR, a Computer Program for Assessing Data Importance to Model Predictions Using Linear Statistics

This document describes the OPR-PPR computer program, which evaluates the importance to model predictions of one or more existing or potential observations, or of increased knowledge about one or more system parameters. OPR-PPR accomplishes this task using linear statistical inference, and reports the results using the Observation-Prediction (OPR) and Parameter-Prediction (PPR) statistics. OPR-PPR is programmed in Fortran 90/95 using modules from the Joint Universal Parameter IdenTification and Evaluation of Reliability (JUPITER) Application Program Interface (API) (Banta and others, 2006). OPR-PPR reads and writes files that are compatible with the JUPITER API data-exchange file formats. OPR-PPR has been developed to be independent of any specific model platform, and can be used with any model that produces the input files necessary for OPR-PPR.