Subduction controls the distribution and fragmentation of Earth’s tectonic plates

Journal name:
Nature
Volume:
535,
Pages:
140–143
Date published:
DOI:
doi:10.1038/nature17992
Received
Accepted
Published online

The theory of plate tectonics describes how the surface of Earth is split into an organized jigsaw of seven large plates1 of similar sizes and a population of smaller plates whose areas follow a fractal distribution2, 3. The reconstruction of global tectonics during the past 200 million years4 suggests that this layout is probably a long-term feature of Earth, but the forces governing it are unknown. Previous studies3, 5, 6, primarily based on the statistical properties of plate distributions, were unable to resolve how the size of the plates is determined by the properties of the lithosphere and the underlying mantle convection. Here we demonstrate that the plate layout of Earth is produced by a dynamic feedback between mantle convection and the strength of the lithosphere. Using three-dimensional spherical models of mantle convection that self-consistently produce the plate size–frequency distribution observed for Earth, we show that subduction geometry drives the tectonic fragmentation that generates plates. The spacing between the slabs controls the layout of large plates, and the stresses caused by the bending of trenches break plates into smaller fragments. Our results explain why the fast evolution in small back-arc plates7, 8 reflects the marked changes in plate motions during times of major reorganizations. Our study opens the way to using convection simulations with plate-like behaviour to unravel how global tectonics and mantle convection are dynamically connected.

At a glance

Figures

  1. Snapshots of convection calculations and of Earth with associated spectral heterogeneity maps of the temperature field and seismic velocity field.
    Figure 1: Snapshots of convection calculations and of Earth with associated spectral heterogeneity maps of the temperature field and seismic velocity field.

    The spectral heterogeneity maps are normalized by the value of the highest power. a, The convection solution with a yield stress of 100 MPa contains a large number of plate boundaries. f, The corresponding spherical harmonic map is dominated by degree 6 in the shallow boundary layer. b, The convection solution with a yield stress of 150 MPa has fewer plate boundaries and a decreasing number of slabs. g, The corresponding spherical harmonic map is dominated by degree 4 at the surface. c, The convection solution with a yield stress of 200 MPa has even fewer plate boundaries. h, The corresponding spherical harmonic map is dominated by degree 4 at the surface. d, The convection solution with a yield stress of 250 MPa has a surface that is barely deformed. i, The corresponding spherical harmonic map is blue and dominated by degree 2. e, ETOPO129 global relief model of Earth and a cross-section through S-wave tomographic model SEMUCB-WM130. j, The corresponding spherical harmonic map of the tomographic model is dominated by degrees 4–5 at the surface. CMB, core–mantle boundary.

  2. Plots of the logarithm of cumulative plate count versus the logarithm of plate size for four yield stress values and Earth.
    Figure 2: Plots of the logarithm of cumulative plate count versus the logarithm of plate size for four yield stress values and Earth.

    The cumulative plate count represents the number of plates that exceed a given area. The graphs contain three data sets for a yield stress of 100 MPa or five data sets for other yield stress values, and the data set for Earth2, in which the distinction between small plates and large plates (indicated by the vertical dashed lines) is around 107.6 km2 (39,800,000 km2). a, Graph for models with a yield stress of 100 MPa, showing a distribution of small and medium plates. b, Graph for models with a yield stress of 150 MPa, showing a distinction between distributions of the large and the small plates. The shift in the distribution occurs at a plate size of about 107.8 km2 (63,100,000 km2). c, Graph for models with a yield stress of 200 MPa, displaying fewer small plates; the groups of small and large plates are distinct and split at about 107.6 km2 (39,800,000 km2). d, Graph for models with a yield stress of 250 MPa, showing only medium and large plates. The division between smaller and larger plates in b and c corresponds to the cross-over of the fitted slopes of the large and smaller plates (Extended Data Fig. 3).

  3. Number of triple junctions per 1,000 km of subduction zones versus the average tortuosity.
    Figure 3: Number of triple junctions per 1,000 km of subduction zones versus the average tortuosity.

    Data are shown for four yield stress values and Earth (see legend). The tortuosity is the ratio of the length of the subduction zone to the length of the great circle between the end points. The error bars represent the standard deviation for each data set.

  4. Global viscosity maps of model 2 and the associated kinematics.
    Figure 4: Global viscosity maps of model 2 and the associated kinematics.

    ac, Maps are separated by 10 Myr. The shapes of the large plates do not change much, whereas the adjustment of the small plates evolves quickly. d, 90 Myr after the first snapshot (a), the distribution of the large plates and smaller plates has evolved substantially. In ad, the top panels show the viscosity of the mantle (colour scale); the bottom panels show the different boundary types (coloured lines) and plate sizes (shading) within the boxed regions in the top panels (which focus on longitudes between −30° and 90° and latitudes between −30° and 30°). The arrows indicate the direction and magnitude (represented by arrow length) of the mantle flow. Plate-size categories are determined in Extended Data Fig. 3.

  5. Maps of the surface of a snapshot from a convection model with a yield stress of 150 MPa and of the plate layout of Earth.
    Extended Data Fig. 1: Maps of the surface of a snapshot from a convection model with a yield stress of 150 MPa and of the plate layout of Earth.

    a, Map of sea-floor age with the youngest ages in red characteristic of mid-ocean ridges and the oldest zones in blue characteristic of subduction zones. m.y., millions of years. b, Map of non-dimensional horizontal divergence, with divergence zones (mid-ocean ridges) shown in red and convergence zones (subduction zones) in blue. c, d, Maps of the plate sizes of the convection model (c) and Earth (d). The plate size categories are determined in Extended Data Fig. 3.

  6. Subsurface temperature of a convection model with a yield stress of 150 MPa showing a diffuse plate boundary.
    Extended Data Fig. 2: Subsurface temperature of a convection model with a yield stress of 150 MPa showing a diffuse plate boundary.

    a, Global temperature (colour scale) and surface velocities (arrows). The dark zones represent subduction zones and the light zones indicate mid-ocean ridges. b, Zoom-in of the red boxed region in a showing a diffuse boundary; the steady lateral change of velocity directions (red arrows) characterizes the intraplate diffuse zone (grey shaded area), allowing the determination of a diffuse boundary (black dashed line).

  7. Plots of the logarithm of the cumulative plate count versus the logarithm of the plate size for the snapshots of model 2 and Earth.
    Extended Data Fig. 3: Plots of the logarithm of the cumulative plate count versus the logarithm of the plate size for the snapshots of model 2 and Earth.

    The data for Earth is taken from ref. 2. The plots show the distribution of microplates in light blue, small plates in mid-blue and large plates in dark blue. The equations of the black fit lines and the correlation coefficients R2 are also shown.

  8. Plot of the fraction of large plates adjoining a triple junction versus the type of triple junction for model 2 and for Earth.
    Extended Data Fig. 4: Plot of the fraction of large plates adjoining a triple junction versus the type of triple junction for model 2 and for Earth.

    The data for Earth is taken from ref. 2. The red rectangles correspond to model 2 and the black circles to Earth. The coloured backgrounds indicate of dominance of each boundary type: blue shows triple junctions that are mainly composed of subduction zones, red shows the dominance of mid-ocean ridges or transform boundaries and green the dominance of diffuse boundaries. T, trenches; R, ridges; D, diffuse boundary. We added a type of triple junction T(RRR); these triple junctions are directly connected to curved trenches and produce back-arc basins with small plates, hence they are included in the area of the plot dominated by subduction zones. The error bars represent the standard deviation of the fraction of large plates around a triple junction for model 2 and Earth.

References

  1. Le Pichon, X. Sea-floor spreading and continental drift. J. Geophys. Res. 73, 36613697 (1968)
  2. Bird, P. An updated digital model of plate boundaries. Geochem. Geophys. Geosyst. 4, 1027 (2003)
  3. Morra, G., Seton, M., Quevedo, L. & Müller, R. D. Organization of the tectonic plates in the last 200 Myr. Earth Planet. Sci. Lett. 373, 93101 (2013)
  4. Seton, M. et al. Global continental and ocean basin reconstructions since 200 Ma. Earth Sci. Rev. 113, 212270 (2012)
  5. Sornette, D. & Pisarenko, V. Fractal plate tectonics. Geophys. Res. Lett. 30, 1105 (2003)
  6. Vallianatos, F. & Sammonds, P. Is plate tectonics a case of non-extensive thermodynamics? Physica A 389, 49894993 (2010)
  7. Sdrolias, M., Roest, W. R. & Müller, R. D. An expression of Philippine Sea plate rotation: the Parece Vela and Shikoku Basins. Tectonophysics 394, 6986 (2004)
  8. Taylor, B., Zellmer, K., Martinez, F. & Goodliffe, A. Sea-floor spreading in the Lau back-arc basin. Earth Planet. Sci. Lett. 144, 3540 (1996)
  9. Moresi, L. & Solomatov, V. Mantle convection with a brittle lithosphere: thoughts on the global tectonic styles of the Earth and Venus. Geophys. J. Int. 133, 669682 (1998)
  10. Trompert, R. & Hansen, U. Mantle convection simulations with rheologies that generate plate-like behaviour. Nature 395, 686689 (1998)
  11. Tackley, P. J. Self-consistent generation of tectonic plates in time-dependent, three-dimensional mantle convection simulations: 1. Pseudoplastic yielding. Geochem.Geophys. Geosyst. 1, 1021 (2000)
  12. Coltice, N., Seton, M., Rolf, T., Müller, R. & Tackley, P. J. Convergence of tectonic reconstructions and mantle convection models for significant fluctuations in seafloor spreading. Earth Planet. Sci. Lett. 383, 92100 (2013)
  13. Ricard, Y., Bercovici, D. & Schubert, G. A two-phase model for compaction and damage: 2. Applications to compaction, deformation, and the role of interfacial surface tension. J. Geophys. Res. 106, 8907 (2001)
  14. Stein, C., Schmalzl, J. & Hansen, U. The effect of rheological parameters on plate behaviour in a self-consistent model of mantle convection. Phys. Earth Planet. Inter. 142, 225255 (2004)
  15. van Heck, H. J. & Tackley, P. J. Planforms of self-consistently generated plates in 3D spherical geometry. Geophys. Res. Lett. 35, L19312 (2008)
  16. Bello, L., Coltice, N., Rolf, T. & Tackley, P. J. On the predictability limit of convection models of the Earth’s mantle. Geochem. Geophys. Geosyst. 15, 23192328 (2014)
  17. Williams, S. E., Müller, R. D. & Landgrebe, T. C. W. An open-source software environment for visualizing and refining plate tectonic reconstructions using high-resolution geological and geophysical data sets. GSA Today 22(4), 49 (2012)
  18. Becker, T. W. & Boschi, L. A comparison of tomographic and geodynamic mantle models. Geochem. Geophys. Geosyst. 3, 1003 (2002)
  19. van Heck, H. J. & Tackley, P. J. Plate tectonics on super-Earths: equally or more likely than on Earth. Earth Planet. Sci. Lett. 310, 252261 (2011)
  20. Tackley, P. J. Modelling compressible mantle convection with large viscosity contrasts in a three-dimensional spherical shell using the yin-yang grid. Phys. Earth Planet. Inter. 171, 718 (2008)
  21. Matthews, K. J., Seton, M. & Müller, R. D. A global-scale plate reorganization event at 105–100 Ma. Earth Planet. Sci. Lett. 355–356, 283298 (2012)
  22. Stegman, D. R., Schellart, W. P. & Freeman, J. Competing influences of plate width and far-field boundary conditions on trench migration and morphology of subducted slabs in the upper mantle. Tectonophysics 483, 4657 (2010)
  23. Barckhausen, U., Ranero, C. R., Cande, S. C., Engels, M. & Weinrebe, W. Birth of an intraoceanic spreading center. Geology 36, 767770 (2008)
  24. Petterson, M. G. et al. Geological–tectonic framework of Solomon Islands, SW Pacific: crustal accretion and growth within an intra-oceanic setting. Tectonophysics 301, 3560 (1999)
  25. Harbert, W. Late Neogene relative motions of the Pacific and North America plates. Tectonics 10, 115 (1991)
  26. Tebbens, S. F. & Cande, S. C. Southeast Pacific tectonic evolution from early Oligocene to present. J. Geophys. Res. 102(B6), 1206112084 (1997)
  27. Cloetingh, S. A. P. L., Gradstein, F. M., Kooi, H., Grant, A. C. & Kaminski, M. M. Plate reorganization: a cause of rapid late Neogene subsidence and sedimentation around the North Atlantic? J. Geol. Soc. Lond. 147, 495506 (1990)
  28. van der Meer, D. G., Torsvik, T. H., Spakman, W., van Hinsbergen, D. J. J. & Amaru, M. L. Intra-Panthalassa Ocean subduction zones revealed by fossil arcs and mantle structure. Nat. Geosci. 5, 215219 (2012)
  29. Amante, C. & Eakins, B. W. ETOPO1 1 Arc-Minute Global Relief Model: Procedures, Data Sources and Analysis www.ngdc.noaa.gov/mgg/global/relief/ETOPO1/image/color_etopo1_ice_low.tif.zip (US National Oceanic and Atmospheric Administration, 2009)
  30. French, S. W. & Romanowicz, B. A. Broad plumes rooted at the base of the Earth’s mantle beneath major hotspots. Nature 525, 9599 (2015)
  31. Lay, T., Hernlund, J. & Buffett, B. A. Core–mantle boundary heat flow. Nat. Geosci. 1, 2532 (2008)
  32. Ricard, Y., Richards, M., Lithgow-Bertelloni, C. & Le Stunff, Y. A geodynamic model of mantle density heterogeneity. J. Geophys. Res. 98, 2189521909 (1993)
  33. Mitrovica, J. X. Haskell [1935] revisited. J. Geophys. Res. 101, 555569 (1996)
  34. Tosi, N. et al. A community benchmark for viscoplastic thermal convection in a 2-D square box. Geochem. Geophys. Geosyst. 16, 21752196 (2015)
  35. Gordon, R. G. in The History and Dynamics of Global Plate Motions (eds Richards, M. A. et al.) 143159 (Geophysical Monograph Series, Vol. 121, American Geophysical Union, 2000)
  36. Goudarzi, M. A., Cocard, M. & Santerre, R. EPC: Matlab software to estimate Euler pole parameters. GPS Solut. 18, 153162 (2014)
  37. Zhong, S., Gurnis, M. & Moresi, L. Role of faults, nonlinear rheology, and viscosity structure in generating plates from instantaneous mantle flow models. J. Geophys. Res. 103, 1525515268 (1998)

Download references

Author information

Affiliations

  1. Laboratoire de Géologie de Lyon, École Normale Supérieure, UMR 5276 CNRS, Université de Lyon 1, 69622 Villeurbanne, France

    • Claire Mallard &
    • Nicolas Coltice
  2. Institut Universitaire de France, 103 Boulevard Saint Michel, 75005 Paris, France

    • Nicolas Coltice
  3. EarthByte Group, School of Geosciences, Madsen Building F09, University of Sydney, New South Wales 2006, Australia

    • Maria Seton &
    • R. Dietmar Müller
  4. Institute of Geophysics, Department of Earth Sciences, ETH Zürich, Sonneggstrasse 5, 8092 Zurich, Switzerland

    • Paul J. Tackley

Contributions

C.M. developed the methodology for analysing the convection models, conducted the plate analysis, contributed to the interpretation and wrote the manuscript. N.C. conducted the convection calculations, contributed to the development of the methodology and analysis, contributed to the interpretation and wrote the manuscript. M.S. and R.D.M. provided guidance with GPlates and scripts, contributed to the interpretation and wrote the manuscript. P.J.T. provided the StagYY convection code, guidance on using it and wrote the manuscript.

Competing financial interests

The authors declare no competing financial interests.

Corresponding author

Correspondence to:

Author details

Extended data figures and tables

Extended Data Figures

  1. Extended Data Figure 1: Maps of the surface of a snapshot from a convection model with a yield stress of 150 MPa and of the plate layout of Earth. (137 KB)

    a, Map of sea-floor age with the youngest ages in red characteristic of mid-ocean ridges and the oldest zones in blue characteristic of subduction zones. m.y., millions of years. b, Map of non-dimensional horizontal divergence, with divergence zones (mid-ocean ridges) shown in red and convergence zones (subduction zones) in blue. c, d, Maps of the plate sizes of the convection model (c) and Earth (d). The plate size categories are determined in Extended Data Fig. 3.

  2. Extended Data Figure 2: Subsurface temperature of a convection model with a yield stress of 150 MPa showing a diffuse plate boundary. (475 KB)

    a, Global temperature (colour scale) and surface velocities (arrows). The dark zones represent subduction zones and the light zones indicate mid-ocean ridges. b, Zoom-in of the red boxed region in a showing a diffuse boundary; the steady lateral change of velocity directions (red arrows) characterizes the intraplate diffuse zone (grey shaded area), allowing the determination of a diffuse boundary (black dashed line).

  3. Extended Data Figure 3: Plots of the logarithm of the cumulative plate count versus the logarithm of the plate size for the snapshots of model 2 and Earth. (226 KB)

    The data for Earth is taken from ref. 2. The plots show the distribution of microplates in light blue, small plates in mid-blue and large plates in dark blue. The equations of the black fit lines and the correlation coefficients R2 are also shown.

  4. Extended Data Figure 4: Plot of the fraction of large plates adjoining a triple junction versus the type of triple junction for model 2 and for Earth. (218 KB)

    The data for Earth is taken from ref. 2. The red rectangles correspond to model 2 and the black circles to Earth. The coloured backgrounds indicate of dominance of each boundary type: blue shows triple junctions that are mainly composed of subduction zones, red shows the dominance of mid-ocean ridges or transform boundaries and green the dominance of diffuse boundaries. T, trenches; R, ridges; D, diffuse boundary. We added a type of triple junction T(RRR); these triple junctions are directly connected to curved trenches and produce back-arc basins with small plates, hence they are included in the area of the plot dominated by subduction zones. The error bars represent the standard deviation of the fraction of large plates around a triple junction for model 2 and Earth.

Additional data