Figure 12.
Thermochemical mantle structure. (a)–(c) Map of mean mantle potential temperature for a mechanically mixed mantle ($\rm T_{pot}$; a) and histograms sorted by tectonic setting for inversions using a fully equilibrated (b) and mechanically mixed mantle (c), respectively. Continental (Q, P, S) and oceanic regions (A, B, C) follow the global tectonic regionalization ‘GTR1’ (Jordan 1981,cf. text for details). In addition, results are shown for hotspots (HS; Anderson & Schramm 2005), subduction zones with a slab in the upper mantle (SZ1) or in the MTZ (SZ2; Hayes et al. 2018) and normal mantle (NM) away from hotspots and slabs. Hotspot locations and slab contours are indicated in (a) by triangles and lines, respectively. (d)–(f) Same as (a)–(c) but for composition, parametrized by basalt fraction (f). (g) Distribution of logarithmic Bayes’ factor (see main text for details) computed from the total misfit values (cf. eq. 8). For a Bayes' factor <0, a mechanical mixture (MM) is favored over an equilibrium assemblage (EA) and vice versa for a Bayes' factor >0.

Thermochemical mantle structure. (a)–(c) Map of mean mantle potential temperature for a mechanically mixed mantle (⁠|$\rm T_{pot}$|⁠; a) and histograms sorted by tectonic setting for inversions using a fully equilibrated (b) and mechanically mixed mantle (c), respectively. Continental (Q, P, S) and oceanic regions (A, B, C) follow the global tectonic regionalization ‘GTR1’ (Jordan 1981,cf. text for details). In addition, results are shown for hotspots (HS; Anderson & Schramm 2005), subduction zones with a slab in the upper mantle (SZ1) or in the MTZ (SZ2; Hayes et al. 2018) and normal mantle (NM) away from hotspots and slabs. Hotspot locations and slab contours are indicated in (a) by triangles and lines, respectively. (d)–(f) Same as (a)–(c) but for composition, parametrized by basalt fraction (f). (g) Distribution of logarithmic Bayes’ factor (see main text for details) computed from the total misfit values (cf. eq. 8). For a Bayes' factor <0, a mechanical mixture (MM) is favored over an equilibrium assemblage (EA) and vice versa for a Bayes' factor >0.

Close
This Feature Is Available To Subscribers Only

Sign In or Create an Account

Close

This PDF is available to Subscribers Only

View Article Abstract & Purchase Options

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

Close