“…In the supplementary information published with the proposed rulemaking for 10 CFR 100 (59 FR 52255), NRC staff cited the need for stability in regulatory standards as a reason for its decision to define the reference probability in DG-1032 as a median value. Additional reasons include the findings that, when median hazard curves were disaggregated, the magnitudes and distances of the controlling earthquakes tended to be more sharply defined and to agree better with the safe shutdown earthquakes of the selected plants than when mean hazard curves were disaggregated (Bernreuter et al 1996).…”