UCLA

This study evaluates the efficacy of two alternative methods for estimating vertical seismic load effects (Ev), which are specified in United States building codes (e.g., ASCE 7-22) and guidelines, and employed in the seismic design of structures. In both approaches, the vertical seismic load is computed as a percentage of the nominal dead load. One method uses 20% of the horizontal short-period design spectral acceleration (SDS) as the dead load factor, while the other computes the dead load factor as 30% of the vertical spectral acceleration value at the (vertical) structural period of interest (Sav). The two approaches are evaluated and compared using (1) the results of probabilistic seismic hazard analysis carried out by the USGS (which is the backbone data for ASCE 7-22) for 19,316 sites in California, and (2) the NGA-West2 ground motion models for over 3,000 seismic scenarios. The results are presented in terms of the probability distribution of the ratio of the Ev values obtained from the two approaches (0.3Sav/0.2SDS) as a function of the vertical period of the structure. The evaluation reveals that the outcomes of applying the two methods are generally not consistent and depend significantly on the vertical period range of the structure. The study pinpoints period ranges where the results of one method substantially diverge from the other and provides recommendations for Ev calculations across various seismic design categories.