The Use and Misuse of Logic Trees in Probabilistic Seismic Hazard Analysis

Abstract

Logic trees have become a standard feature of probabilistic seismic hazard analyses (PSHA) for determining design ground motions. A logic tree's purpose is to capture and quantify the epistemic uncertainty associated with the inputs to PSHA and thus enable estimation of the resulting uncertainty in the hazard. There are many potential pitfalls in setting up a logic tree for PSHA, mainly related to the fact that in practice, it is questionable that the requirements that the logic-tree branches be both mutually exclusive and collectively exhaustive can actually be met. Careful consideration is also required for making use of the output; in particular, in view of how PSHA is employed in current engineering design practice, it may be more rational to determine the mean ground motion at the selected design return period rather than to find the ground motion at the mean value of this return period.