Abstract Geological hiatus maps provide a proxy for dynamic topography spanning back approximately 165 Myr, providing a new data set to test global mantle circulation models (MCMs). Here, we show how these maps can be generated synthetically from the dynamic topography predictions of an MCM and used for model assessment. Filtering the model predictions into an equivalent synthetic data set is required prior to a direct comparison to the observed maps—analogous to the tomographic filtering of MCM outputs. The dynamic topography field of our reference MCM has overestimated amplitudes at the longest wavelengths (l≤4) by a factor of approximately two, which is a common property of geodynamic models. We find that this in turn results in synthetic hiatus maps with overestimated hiatus areas in Africa and Europe, as well as slower temporal changes than the observed maps. Altering the dynamic topography spectrum to match present-day observations significantly improves the match to the observed hiatus maps. We therefore find that the synthetic hiatus maps can be used to assess the past dynamic topography predictions of MCMs, and that a model which better fits the present-day spectrum of dynamic topography also provides better predictions in the past, as expected.
Brown et al. (Thu,) studied this question.