Research

Global plate tectonic reconstructions

Building internally consistent models of where every plate boundary sat, from the present day back through a billion years of tectonic history, with the open-source GPlates software and the EarthByte Group.

Müller et al. (2016, AREPS) pulled together decades of regional plate model work into the first internally consistent global model of every plate boundary spanning the last 200 Myr, built as a continuously evolving topological network rather than a fixed set of static polygons. Comparing rates and directions of separation across key plate pairs along representative tectonic flowlines (Figure 8) makes it possible to tell which changes in spreading behaviour are genuine global tectonic events, and which are just artefacts of individual regional models.

Globes and time-series plots showing divergence rate and direction along eight key tectonic flowlines, including North America-Greenland, Africa-South America, Africa-Antarctica, Australia-Antarctica, Pacific-Farallon and India relative to Madagascar, Antarctica and Australia

Figure 8, Müller et al. (2016).

Merdith et al. (2017, Gondwana Research) pushed a full-plate model back through the Neoproterozoic for the first time, linking the breakup of Rodinia to the assembly of Gondwana using geological constraints on individual terranes rather than paleomagnetic poles alone. Merdith et al. (2021, Earth-Science Reviews) then joined that Neoproterozoic model smoothly onto the younger Phanerozoic plate model, producing the first continuous, self-consistent full-plate reconstruction spanning a billion years of Earth history. Figure 17, from the 2021 paper, shows how the Arabian-Nubian Shield and the Azania and TOAST terranes were progressively incorporated into Gondwana between 800 and 520 Ma.

Two globes showing the amalgamation of the Arabian-Nubian Shield, Azania and the TOAST terrane into Gondwana at 800 Ma and 520 Ma

Figure 17, Merdith et al. (2021).

Reconstructing seafloor age is straightforward where the ocean floor still exists, but most seafloor older than about 180 Myr has already been subducted. Williams et al. (2021, Geoscience Frontiers) built an efficient workflow for reconstructing that missing seafloor age structure within a full-plate model, and used it to explore a range of plausible age distributions in ocean basins that no longer exist. The animation below steps a resulting seafloor age grid forward through geological time, generated automatically from a topological plate reconstruction.

Animated seafloor age grid evolving through geological time

An automatically generated seafloor age grid stepping through geological time, from agegrid-0.1.

Treating every plate as perfectly rigid is a convenient simplification, but real continental margins stretch, rift and shorten. Müller et al. (2019, Tectonics) built the first global plate model to explicitly include that deformation, capturing the progressive extension of continental margins since rifting began within Pangea around 240 Ma, along with major failed rifts and zones of continental collision. The snapshot below, from 30 Ma, shows accumulated stretching (blue) and shortening (red) along plate boundaries and deforming continental interiors worldwide.

Global map at 30 Ma showing plate boundaries and accumulated continental extension (blue) and shortening (red) across deforming plate margins

Deforming plate model at 30 Ma, Müller et al. (2019).