Uncertainty of Tectonic Reconstructions by Plate ID
Ben R. Mather - EarthByte Group, The University of Sydney
Software Used:
- pyGPlates rev. 22
Model Setup:
We seed a bunch of points on the sphere using the legendary spherical meshing package, stripy (https://www.github.com/underworldcode/stripy), that remain static through time as we query the plate ID at each lat/lon coordinate through time using pyGPlates. Iterating through a wide selection of plate reconstructions, we determine where on the sphere the plate IDs match and where they do not. A score is kept for how many reconstructions have contradicting plate IDs - i.e. if a point at 142,35 (lon/lat) has a different plate ID for 3 reconstructions, then it is assigned an uncertainty rating of 3.
Conditions:
| Rotation files | Year | Published | Valid range |
|---|---|---|---|
| Global EarthByte model | 2019 | Tectonics | 250 - 0 Ma |
| Matthews et al. | 2016 | Global and Planetary Change | 410 - 0 Ma |
| Muller et al. (AREPS) | 2015 | Annual Reviews of Earth and Planetary Science | 230 - 0 Ma |
| Shephard et al. | 2013 | Earth Science Reviews | 200 - 0 Ma |
| Seton et al. | 2012 | Earth Science Reviews | 200 - 0 Ma |
| Gurnis et al. (Caltech) | 2012 | Computers & Geoscience | 140 - 0 Ma |
Results:
It is not unexpected that the greatest different rotation models occurs at plate boundaries. Also, it is clear that the plate IDs assigned to certain continents have changed as plate reconstructions are updated. A caveat is that rotations of the plates with respect to the absolute reference frame are not captured, just the region occupied by a plate. Nonetheless this is a useful tool to compare how a plate reconstruction stacks up against others. It is a good validation experiment to examine whether any new plate reconstruction is wildly different to those that came before.