Ground Motion Uncertainty in Deriving the Empirical Fragility Functions after the Feb. 6 2023 Kahramanmaraş-Türkiye Earthquakes

Empirical methods for the development of fragility functions employ observational damage data collected after an earthquake. Observations at various locations, which are graded based on a predefined damage scale, are correlated to a ground motion intensity measure and as a result of this statistical process fragility functions are generated. This procedure essentially requires characterization of corresponding ground motion intensity levels that the buildings in the damage data set have experienced. Ideally, recorded ground motion data across the surveyed areas would be used for intensity level assignments. However, due to the scarcity of ground motion recordings, ground motion intensity values over the geographical extent where the damaged buildings spread out are obtained, in practice, through ground motion prediction equations (GMPEs) or physics based ground motion simulations, which come with certain complexities and at additional computational cost. The estimated ground motions can then be further improved by the incorporation of the recorded values, if any available. After the Feb. 6, 2023 Kahramanmaraş-Türkiye earthquakes we derived fragility functions (Hancilar and Acikgoz, 2024a) using the official field based damage data (2023) and the rapid ground shaking estimations (Hancilar et al., 2023). We recently revisited our analyses for the computation of the spatial distributions of ground shaking intensities by incorporating different local site effect models, ground motion predictive models as well as by implementing bias adjustments on the ground motion estimations with the addition of more strong motion recording stations (Hancilar and Acikgoz, 2024b). This study deals with the re-derivation of fragility functions with different ground motion inputs while the building damage data kept unchanged. The resulting fragilities are compared and the effect of the ground motion uncertainty is examined.