Unrecognized Sources of Uncertainties (USU) in Experimental Nuclear Data
Evaluated nuclear data uncertainties are often perceived as unrealistic, most often because they are thought to be too small. The impact of this issue in applied nuclear science has been discussed widely in recent years. Commonly suggested causes are: poor estimates of specific error components, neglect of uncertainty correlations, and overlooked known error sources. However, instances have been reported where very careful, objective assessments of all known error sources have been made with realistic error magnitudes and correlations provided, yet the resulting evaluated uncertainties still appear to be inconsistent with observed scatter of predicted mean values. These discrepancies might be attributed to significant unrecognized sources of uncertainty (USU) that limit the accuracy to which these physical quantities can be determined. The objective of our work has been to develop procedures for revealing and including USU estimates in nuclear data evaluations involving experimental input data. We conclude that the presence of USU may be revealed, and estimates of magnitudes made, through quantitative analyses. This paper identifies several specific clues that can be explored by evaluators in identifying the existence of USU. It then describes numerical procedures to generate quantitative estimates of USU magnitudes. Key requirements for these procedures to be viable are that sufficient numbers of data points be available, for statistical reasons, and that additional supporting information about the measurements be provided by the experimenters. Realistic examples are described to illustrate these procedures and demonstrate their outcomes as well as limitations. Our work strongly supports the view that USU is an important issue in nuclear data evaluation, with significant consequences for applications, and that this topic warrants further investigation by the nuclear science community.