Supplementary Material for Circum-Arctic mantle structure and long-wavelength topography since the Jurassic Shephard, G. E., Flament, N., Williams, S., Seton, M., Gurnis, M., and Müller, R.D. Supplementary Methods Absolute reference frames and Net Lithospheric Rotation (NLR) The hybrid absolute plate reference frame of Seton et al. (2012) (case C2) is based on a moving Indian/Atlantic hotspot model (O’Neill et al., 2005) for times younger than 100 Ma and on a True Polar Wander (TPW)-corrected palaeomagnetic model (Steinberger and Torsvik, 2008) for older times (Table 2). The use of the hybrid absolute plate motion of O’Neill et al. (2005) and Steinberger and Torsvik (2008) implies NLR in excess of 0.4°/Myr between ~40-60, 65-80, 110-115 and 180-215 Ma (Figure S13). NLR is large at present-day in a Pacific hotspot reference frame (~ 0.44°/Myr HS3, Conrad and Behn, 2010) and geologically recent NLR (since ~ 50 Ma) has an overall westward direction with estimated rates previously 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
34
Embed
authors.library.caltech.eduauthors.library.caltech.edu/53454/...Material_Shephard… · Web viewSupplementary Material for. Circum-Arctic. mantle structure and long-wavelength topography
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Supplementary Material for
Circum-Arctic mantle structure and long-wavelength topography since the
Jurassic
Shephard, G. E., Flament, N., Williams, S., Seton, M., Gurnis, M., and Müller, R.D.
Supplementary Methods
Absolute reference frames and Net Lithospheric Rotation (NLR)
The hybrid absolute plate reference frame of Seton et al. (2012) (case C2) is
based on a moving Indian/Atlantic hotspot model (O’Neill et al., 2005) for times
younger than 100 Ma and on a True Polar Wander (TPW)-corrected
palaeomagnetic model (Steinberger and Torsvik, 2008) for older times (Table 2).
The use of the hybrid absolute plate motion of O’Neill et al. (2005) and
Steinberger and Torsvik (2008) implies NLR in excess of 0.4°/Myr between ~40-
60, 65-80, 110-115 and 180-215 Ma (Figure S13). NLR is large at present-day in
a Pacific hotspot reference frame (~ 0.44°/Myr HS3, Conrad and Behn, 2010)
and geologically recent NLR (since ~ 50 Ma) has an overall westward direction
with estimated rates previously ranging between 1.5-9 cm/year (or
~0.11+_0.03°/Myr) depending on reconstructions (e.g. Ricard et al., 1991;
Becker, 2006; Torsvik et al., 2010). While a component of NLR throughout time
may be real, increasing uncertainty of the plate reconstruction back in time may
result in unrealistically large NLR, especially concerning the velocities of the
large plates comprising Panthalassa. Our plate models are constructed with
continuously closing plates (Gurnis et al., 2012), which allow us to define global
surface velocity fields through time and to calculate the NLR implied by a given
reference frame as in Torsvik et al. (2010) and Alisic et al. (2012).
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
We used three approaches to minimize NLR from the absolute reference frame in
our geodynamic model cases (Tables 2 and S3); (i) computing and removing the
NLR from the plate reconstruction (case C3), (ii) including a low-viscosity
asthenosphere to decouple the lithosphere from the sub-asthenospheric mantle
(cases C2 and C8) and (iii) changing the absolute reference frame by using the
finite rotations from the moving hotspot reference frame of Torsvik et al. (2008)
rather than that of O’Neill et al. (2005) for ages younger than 70 Ma (models C1,
C4, C5 as well as C6-C8). We also change the absolute motion of the Pacific during
the Cenozoic by changing the poles of rotation between east and west Antarctica
(from Cande et al. [2000] to Granot et al. [2013]). The resultant NLR for this
latter absolute plate motion model (C1, C4-C8) is < 0.4°/Myr for all times,
although it is slightly more elevated for the last 20 Ma (~ 0.18°/Myr) than the
previous reference frame (O’Neill et al., 2005; Steinberger and Torsvik, 2008:
~0.12°/Myr; C2, Table 2, Fig. S13).
In addition to differences in the absolute reference frame we explored different
mantle parameters including viscosity profile, initial slab depth, slab dip and
basal layer density (Figs. S3, S4, S6-8, S10-15, Table S3). We find that changes in
dynamic topography are small and do not affect our main conclusions. Rates of
dynamic topography for alternative cases C3-C5 are usually in the order of ±5
m/Myr from those of C1 (Table S1, S2), and are compatible with the geological
constraints presented in the main text (see Figs. 7e-h). Notably, under Eurasia,
alternative cases C3-C5 (Figs. S10) predict a similar two-slab configuration
(Mongol-Okhotsk slab to the west and north-eastern Panthalassa slab to the
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
east) to those of C1 but with locations offset by ± 5° longitude (10° to the east for
slab (m) in C5, though the use of a depth-dependent viscosity was not ideal, see
below). Case C4, illustrates that increasing the slab dip does not significantly
change the results.
In addition to C1-C5 and in the interests of illustrating the main suite of
parameters tested (see also Flament et al., 2014) we present an extended set of
eight cases (C1-C8) in Figs. S14 and S15. These figures illustrate our investigation
of the effect of rheological parameters on lower mantle structure and our
selection of a set of parameters for C1 by visual comparison between predicted
mantle temperature and seismic tomography along arbitrary cross-sections. For
example, C6 and C8 both include a linear increase in viscosity for the lower
mantle; looking under Eurasia (Fig. S15) in case C6, which has a higher density
basal layer, the predicted volume of slabs is systematically too small, whereas in
case C8, which has a lower density basal layer and asthenosphere, slabs are
significantly offset compared to seismic tomography. C7, which also has a lower
basal density over-predicts the amount of slab material and has dominant
upwellings. Under North America (50°N, Figure S14) the alternative cases are
similar to each other and to seismic tomography (no qualitatively “best” case,
though C8 seems to under-predict slab volumes at this location). We therefore
opted for a dense basal layer and a layered viscosity structure, with no depth-
dependent viscosity in the lower mantle for reference case C1. Note that the
influence of alternative parameters on the pattern of dynamic topography (right
panels of Figs. S14 and S15) is small; our main conclusions are largely unaffected
by parameter selection.
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
Supplementary References
Alisic, L., Gurnis, M., Stadler, G., Burstedde, C., and Ghattas, O., 2012, Multi-scale
dynamics and rheology of mantle flow with plates. Journal of Geophysical
Research, v.117 doi:10.29/2012JB009234
Ballance, P.F., 1993, in South Pacific Sedimentary Basins v.2 of Sedimentary
Basins of the Word P.F., Balance (Ed) Elsevier Amsterdam p.93-110.
Becker, T.W., 2006, On the effect of temperature and strain-rate dependent
viscosity on global mantle flow, net rotation, and plate-driving forces.
Geophysical Journal International v.167 p.943-957.