Polar wander leads to large differences in past climate reconstructions

Abstract

Abstract The position of continents and oceans in geological time dictates the biogeographic dispersal of life, influences the preservation of mineral resources, and informs our understanding of Earth’s climate trajectory. Reconstructing past crustal locations requires a plate tectonic model that differs depending on whether the model uses a mantle reference frame, or a paleomagnetic reference frame, which considers the rotation between the plate-mantle system and the planetary core (true polar wander). However, reconstructions with mantle or paleomagnetic reference frames have been used interchangeably without quantifying their impact on paleoclimate simulations. Here, we conduct a sensitivity experiment to assess the impact of using a mantle versus a pure paleomagnetic absolute reference frame in simulating global paleoclimate. We show that throughout the Mesozoic and Cenozoic, reference frame choice leads to regional surface temperature differences of over 20 °C and 15 °C respectively. An analysis of Jurassic glendonite and Eocene crocodilian distribution suggest favourable conditions that are more consistent with a paleomagnetic reference frame. These results emphasize the importance of selecting an appropriate tectonic reference frame in deep-time climate research. We advocate for adopting a pure paleomagnetic reference frame, which provides a more reliable record of paleo-latitudes by capturing motions from true polar wander.