Peak over threshold vis-à-vis equivalent triangular storm: Return value sensitivity to storm threshold

The objective of this paper is to compare two methodologies adopted in the context of long-terms statistics: Equivalent Triangular Storm (ETS) and Peak Over Threshold (POT), applied both with the Goda model and with the Generalized Pareto Distribution. Both methodologies are currently used for conducting long-term statistics oriented to the determination of relevant input parameters adopted in the design of coastal structures. The paper proposes a dissemination of the key characteristics of the methods and elucidates the fundamental steps necessary for mechanizing the methods starting from available wave data. Then, it discusses numerical results pertaining to both Mediterranean and Ocean data. In this context, the paper focuses on the stability of the methods with respect to the (subjectively) selected storm thresholds. It shows that the ETS method is more stable than the POT. Results pertain to both omnidirectional and directional return values.