Digital Paleoclimate: Integration, Simulation, and Assimilation

Instrumental observations, mainly starting after industrial revolution, is usually the primary basis for understanding climate dynamics, building state of art models, and projecting the future. However, the length of observation data, i.e., approximately a hundred of years, fails to cover several multi-decadal-, centennial-, or even longer-timescale climate variability, which significantly restricts the ability of climate models on long-term predictions. From this perspective, paleoclimate reconstructions (e.g., tree ring, coral, Tridacna, stalagmites, loess, lake and ocean sediments), which give clear vision for these multi-scale climate variability and beneficially supplement to instrumental data. In particular, the paleoclimate data can provide unique and irreplaceable constraints for climate models.

In the past decades, various high-quality paleoclimate records have been accumulated and scientists start to attempt the data integration, aiming to establish a numerical paleoclimate dataset like modern grid network. Meanwhile, long-term transient experiments by earth system models, spanning the past tens of thousands of years, also become feasible with rapid development of high performance computing. It is worth noting that the combination of paleoclimate records and simulation outputs using assimilation methods will lead to milestone advances for comprehensively understanding multi-scale Earth’s climate evolution. Furthermore, artificial intelligence technologies have been gradually applied in paleoclimate studies, which may also be a new and cutting-edge direction. In this regard, it is evident that paleoclimate studies are now transforming from traditional geological paradigm to digital paradigm.

The purpose of this Collection is to push forward digital transformation of paleoclimate studies. We welcome Original Research and Review articles as well as Comment and Perspective addressing the Earth’s multi-scale paleoclimate changes, such as the paleo-data integrations, paleoclimate simulations, and data assimilation, along with the applications of machine learning algorithms in paleoclimate studies.