Abstract Understanding the history of Earth rotation variations and its connection to mantle dynamics is one of the most important problems is global geophysics. However, our knowledge of these variations—in particular those induced by climate on geological timescales—is limited due to both modeling deficiencies and the scarcity of paleoclimate data. In order to advance our understanding of this problem, here we first develop a new probabilistic deep learning methodology called Physics‐Informed Diffusion Model (PIDM). We then use PIDM in conjunction with the recently available paleoclimate data—specifically, sea level variations since the Late Pliocene—to precisely reconstruct the history of climate‐induced changes in the Earth's rotation rate (i.e., Length of Day variations: LOD). We reconcile LOD inferred from various climate models and paleoclimate proxies (i.e., geological records such as fossil benthic foraminifera and coral reefs). Based on our reconstructions of LOD, we unravel (a) large‐amplitude fluctuations due to Quaternary ice ages, surpassing the magnitude of the currently known processes including those of atmosphere, land hydrology, and core, (b) a previously unrecognized secular trend due to changes in the Earth's oblateness caused by the outset of Northern Hemispheric ice sheets, and (c) the almost unprecedented rate of increase in the length of day caused by century climate change.
Shahvandi et al. (Sun,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: