On the Inner‐Core Differential‐Rotation (Un)Resolvability From Earthquake Doublets: The Traps of Data Selection

Abstract

AbstractThe phenomenon of differential rotation of the Earth's inner core relative to the mantle is a subject of interest in geodynamo modeling that has been validated by seismological observations, mainly via the earthquake‐doublets method. Although recent studies converge on the time‐varying differential rotation of the inner core relative to the mantle, favoring a decadal variation, the inferred models significantly differ. Here, considering the data selection, the observed data structure, and the subjective model parameterizations, which we avoid by employing a Bayesian transdimensional approach, we show that the recent best‐fit model by Yang and Song (2023, https://doi.org/10.1038/s41561‐022‐01112‐z) featuring the 70‐year decadal variation is not obtained when all available data are considered. Namely, including only a small number of discarded earthquake doublets (<10%) changes the period of the inner‐core differential rotation fluctuation to 20–30 years. More earthquake‐doublet data are required to address the non‐uniqueness of the inversion problem.