Skip to content

Validation

This tab lists the tests that compare Aragog with an independent result: a published benchmark, an analytical limit or a second implementation. Each case in the table carries the reference_pinned marker. Where a page section exists, the table links it.

  • Core: the core module against Nimmo (2015)1617, the leeds and leeds_thermal models of thermal_history (Greenwood et al. (2021)12) and analytical limits.
  • Core alloy liquidus: the liquidus of iron with S, C and O. At high pressure it is checked against the liquidus curves and eutectic points of Mori et al. (20176), Mashino et al. (20195) and Oka et al. (20198). From 1 bar to 25 GPa the eutectic is checked against Fei et al. (19974) and 12 eutectic check points (measured points, fit values and bounds), and the liquidus against Buono & Walker (20112) and Chen et al. (20083). The page also checks the rule for several light elements, the slope of the liquidus against pressure and the 2 reads of every curve taken from a figure.
  • Tabulated core: the core on the PALEOS liquid iron isentrope against the Gaussian budget of a table with constant properties, the Earth budget of Nimmo (2015)16, an independent integral of the adiabatic gradient, the slope of the pure-iron melting curve, the per-call core ledger and the ab initio entropy of fusion of GonzΓ‘lez-Cataldo & Militzer (2023)22.
  • Surface boundary: the prescribed surface temperature (outer_boundary_condition = 5) against closed-form conduction solutions and a grid-convergence study.
  • Surface lid: the sub-grid surface lid (subgrid_lid) against the freezing of a lava layer, a cooling half-space, a steady crust and a fine mesh, for a solid shell and a shell that freezes from the mush, the mesh and floor dependence of the lid base at the melt temperature, and the lid surface and onset weight under an atmosphere.
  • Phase transitions: the blend of the material properties at the solidus and the liquidus against the blend formula of SPIDER, and the NumPy and JAX right-hand sides inside the blend.
  • Mesh refinement: the mesh refined towards the surface and the CMB against its set end cells and against finer uniform meshes, and the uniform mesh against its stored results.
  • Numerics: the first-principles checks of the solver, the phase-boundary rate controller, the equation of state and the NumPy and JAX right-hand sides, including the reference_pinned solid-rheology solver-wiring benchmarks.
  • Solid-state convection: the saturating integral activation enthalpy, the viscous mixing-length top-slope benchmarks, the 0-D stagnant-lid comparison and the internal-heating scaling exponents. The Arrhenius creep, plastic yield and stagnant-lid closure have worked numbers on the Explanation page. The reference_pinned tests of the Arrhenius creep and the yield stress are in the table below; the stagnant-lid closure has unit tests only.

Cases by process

Process Case Test Reference Tolerance Page
Core stable layer Young layer against the ierfc solution tests/test_core_layer.py::test_a_young_layer_follows_the_erfc_solution_of_a_half_space constant-flux half space (analytical) 1 % of the expected rise of the top cell Core
Core stable layer Quasi-static depth of a steady layer tests/test_core_layer.py::test_a_steady_layer_reaches_the_quasi_static_depth conducted adiabatic flow equal to the CMB flow 5 % Core
Core stable layer Layers under fixed and eroding CMB flows tests/test_core_verification_page.py::test_the_stable_layer_meets_the_thermal_history_bounds leeds_thermal (Greenwood et al. (2021)12) T_cmb 5 K, T_cen 10 K, layer depth within a factor 1.5, inner core 2 % at the end, onset 5 % Core
Core alloy liquidus Fe-S liquidus curves tests/test_core_alloy_liquidus.py::test_the_fe_s_liquidus_curves_of_mori_are_the_ideal_form_on_fe3s_units Mori et al. (20176), Fig. 9 RMS 5 K (first read), 6 K (second) Core alloy liquidus
Core alloy liquidus Fe-C liquidus curves tests/test_core_alloy_liquidus.py::test_the_fe_c_misfit_against_mashino_stays_as_stated Mashino et al. (20195), Fig. 8 stated RMS + 1 K Core alloy liquidus
Core alloy liquidus Eutectic temperatures tests/test_core_alloy_liquidus.py::test_the_eutectic_temperatures_meet_the_papers, ::test_the_fe_o_eutectic_temperature_meets_the_second_read Mori et al. (20176), Mashino et al. (20195), Oka et al. (20198) 50 K (fits), 0.354 pt (second read) Core alloy liquidus
Core alloy liquidus Several light elements tests/test_core_alloy_liquidus.py::test_the_harmonic_rule_gives_the_stated_depressions rules written out by hand 0.1 K Core alloy liquidus
Core alloy liquidus Low-pressure eutectic tests/test_core_alloy_liquidus_low_pressure.py::test_the_fe_s_eutectic_nodes_are_the_mean_of_two_reads_of_fei_1997, ::test_the_fe_c_eutectic_below_5_gpa_is_the_monotone_cubic, ::test_the_low_pressure_eutectic_against_the_check_points Fei et al. (19974), Fig. 2; Nakajima et al. (20097), p. 207; 12 eutectic points from 1 bar to 25 GPa relative 1e-12 (nodes, cubic); stated misses to 0.02 K and 0.002 wt% Core alloy liquidus
Core alloy liquidus Fe-S liquidus from 1 bar to 15 GPa tests/test_core_alloy_liquidus_low_pressure.py::test_the_low_pressure_liquidus_misses_the_measured_points_as_stated, ::test_the_two_reads_of_the_1_bar_points_agree, ::test_from_10_to_14_gpa_the_liquidus_falls_from_a_higher_s_content_than_the_data Buono & Walker (20112), Fig. 6 and Table 1; Chen et al. (20083), Fig. 1 RMS and mean to 0.05 K; reads to 0.5 pt, 0.28 wt% and 3.2 K Core alloy liquidus
Core alloy liquidus 2 reads of each figure curve tests/test_core_alloy_liquidus.py::test_the_two_reads_of_each_figure_curve_agree first read (vector path) half the drawn line width Core alloy liquidus
Tabulated core Constant-property table against the Gaussian budget tests/test_core_budget_tabulated.py::test_a_constant_property_table_gives_the_gaussian_budget the Gaussian adiabat (analytical) 1e-6 relative Tabulated core
Tabulated core Nimmo Earth budget on the PALEOS isentrope tests/test_core_tabulated_nimmo.py::test_the_nimmo_earth_budget_on_the_paleos_isentrope Nimmo (2015)16, ch. 8.02, Tables 2 and 4 Q_L/Q_g 1e-6 against the Gaussian path; PALEOS terms within wide bounds Tabulated core
Tabulated core Isentrope against the integral of the adiabatic gradient tests/test_core_tabulated_validation.py::test_the_isentrope_equals_the_integral_of_nabla_ad the same table, independent column 1e-5 in T Tabulated core
Tabulated core Freezing direction of pure iron tests/test_core_tabulated_validation.py::test_a_pure_iron_core_freezes_from_the_centre Anzellini et al. (2013)21 melting curve margin above 0.05 Tabulated core
Tabulated core Per-call core ledger tests/test_core_tabulated_ledger.py::test_the_tabulated_core_ledger_converges_with_the_tolerance energy conservation, the rule of the Gaussian path calls 1 to 3 fall 20 times to below 1e-5 at 1000 times tighter tolerances; calls 4 and 5 below 2e-6 Tabulated core
Tabulated core Entropy of fusion tests/test_core_tabulated_validation.py::test_the_entropy_of_fusion_against_the_ab_initio_value GonzΓ‘lez-Cataldo & Militzer (2023)22, p. 9 ratio 1.105 pinned to 5e-4; no bound on the agreement Tabulated core
Core energy budget Earth budget tests/test_core_verification_page.py::test_the_earth_budget_meets_nimmo_tables_4_and_5 Nimmo (2015)16, Tables 4 and 5 2 %, four documented exceptions Core
Core energy budget Printed values of the two Nimmo core models tests/test_core_nimmo_benchmarks.py (the reference_pinned tests except the cross-check) Nimmo (2015)1617, ch. 9.08 Table 2, Figure 3 caption (p. 211) and p. 212; ch. 8.02 Table 2 5e-4 to 3 % per printed value, Cr 5 % and 15 %, the baseline headline 5 % to 45 % Core
Core energy budget Model-2 budget terms at 4180 K tests/test_core_nimmo_benchmarks.py::test_budget_terms_match_thermal_history_cross_check leeds (Greenwood et al. (2021)12); Nimmo (2015)16, ch. 8.02 Tables 2 and 4 1e-5 to 2 % per term Core
Core energy budget Budget terms tests/test_core_verification_page.py::test_budget_terms_match_the_thermal_history_table leeds (Greenwood et al. (2021)12) 1e-3 relative Core
Core energy budget Entropy terms tests/test_core_entropy.py::test_entropy_capacities_match_thermal_history leeds (Greenwood et al. (2021)12) 1e-6 to 1e-4 relative Core
Core dynamo Efficiency factors and dynamo threshold tests/test_core_entropy.py::test_chr09_efficiency_factors_reproduce_printed_values, ::test_dynamo_threshold_and_margin Christensen et al. (2009)10; Nimmo (2015)16 1e-3 relative; threshold below 6.5 TW Core
Core thermal history Core history tests/test_core_verification_page.py::test_core_history_matches_the_thermal_history_table leeds (Greenwood et al. (2021)12) T_cmb 0.2 K, T_cen 1 K, ICB radius 1 %, the same onset Core
Core numerics NumPy and JAX right-hand sides tests/test_jax_dsdt_core_module.py::test_rhs_parity_with_numpy_on_driven_state NumPy implementation 1e-8 relative Core
Surface boundary Steady heated shell tests/test_entropy_bc5_validation.py::test_the_steady_heated_shell_matches_the_closed_form Turcotte and Schubert (2014)1 Eq. 4.40 surface flux 1e-8 relative; temperature order 2.0 Β± 0.2; NΒ² times the top-cell error 0.159 of the rise Β± 10 % (a pin of the scheme, measured) Surface boundary
Surface boundary Cooling below a fixed surface temperature tests/test_entropy_bc5_validation.py::test_cooling_below_a_fixed_surface_temperature_follows_the_half_space_flux Turcotte and Schubert (2014)1 Eq. 4.116 1 % at every output time; signed error within 5e-4 of the grid and curvature estimates Surface boundary
Surface boundary Grid convergence with convection tests/test_entropy_bc5_validation.py::test_the_convecting_shell_converges_with_second_order convergence order and Richardson extrapolation order 2.0 Β± 0.4; 1e-3 of the rise from the extrapolation Surface boundary
Surface lid Crust growth over melt tests/test_surface_lid_neumann.py::test_v1_crust_growth_follows_neumann Turcotte and Schubert (2014)1 Eqs. 4.136 and 4.141 1 % (also 1e-8) at three times for St = 0.1, 1, 3.381 Surface lid
Surface lid Cooling solid without latent heat tests/test_surface_lid_validation.py::test_v2_half_space_cooling Turcotte and Schubert (2014)1 Eq. 4.116 1 % at every output time Surface lid
Surface lid Steady crust under a fixed base flux tests/test_surface_lid_validation.py::test_v3_quasi_steady_crust energy balance of a steady conductive crust 1e-4 relative after ten relaxation times Surface lid
Surface lid Cooling solid shell on a fine mesh tests/test_surface_lid_v2_grid.py::test_v2_lid_matches_the_resolved_mesh 4000-node solver run without the lid 1 % at every output time Surface lid
Surface lid Shell freezing from the mush on a fine mesh tests/test_surface_lid_v2_grid.py::test_v2_melt_lid_matches_the_resolved_mesh 4000-node solver run without the lid 2 % at every output time, flux and crust Surface lid
Surface lid Molten shell freezing through: hand-over of the lid and the coarse mesh after it tests/test_surface_lid_v2_grid.py::test_c3_molten_lid_matches_the_resolved_mesh 800-node solver run without the lid, top cell 0.125 m 3 % at every output time Surface lid
Surface lid State on both sides of a hand-over tests/test_surface_lid_v8_merge_run.py::test_v8_restart_over_the_merge_gives_the_stored_merge_state stored run of the same column on 100 nodes 1e-6 relative; needs ARAGOG_LOWT_EOS_DIR Surface lid
Surface lid Stagnant-lid benchmark rows with the lid tests/test_surface_lid_v5_rows.py::test_row_keeps_its_values_with_the_lid the same row without the lid 1e-7 relative in each observable; hand-over time 1e-4 of the stored run Surface lid
Surface lid Lid surface under an atmosphere (Robin surface) tests/test_surface_lid_atmosphere.py::test_the_robin_surface_solves_the_energy_and_the_surface_balance surface balance and profile energy for constant heat capacity (closed form) 1e-9 relative on T_s and flux, NumPy and JAX Surface lid
Surface lid Weight of the lid at its onset tests/test_surface_lid_atmosphere.py::test_the_onset_weight_rises_without_a_step_below_the_base tanh form in the steady crust thickness (closed form) 1e-12 relative, NumPy and JAX Surface lid
Phase transitions Properties on the solidus and the liquidus tests/test_matprop_smooth_width_validation.py::test_on_each_boundary_every_property_is_the_mean_of_the_two_branches SPIDER get_smoothing (util.c): weight 1/2 on each boundary 1e-12 relative Phase transitions
Mesh End cells of a refined mesh tests/test_mesh_refinement_validation.py::test_the_refined_mesh_has_the_set_end_cells the end-cell thickness set in the config 1e-6 relative (radius mode), 2 mm (mass-coordinate mode) Mesh refinement
Mesh Results against the top cell tests/test_mesh_refinement_study.py::test_the_results_change_as_the_top_cell_gets_thinner stored study runs; spreads against a perturbed pair each quoted number to its digits Mesh refinement
Mesh Long run and tables tests/test_mesh_refinement_study.py::test_the_long_run_and_the_tables stored runs; PALEOS liquid table, Zenodo 19680050 each quoted number to its digits Mesh refinement
Mesh Coarsened CMB cell with the core module tests/test_mesh_refinement_study.py::test_a_coarsened_cmb_cell_changes_the_core_results stored runs against a perturbed copy each quoted number to its digits Mesh refinement
Mesh Default CMB cell against finer uniform meshes tests/test_core_mesh_study.py::test_the_default_cmb_cell_against_finer_uniform_meshes the uniform 640-node mesh (U640) of the stored study runs each quoted number to its digits Mesh refinement
Mesh Mesh without refinement tests/test_mesh_refinement_off_reference.py stored arrays of the uniform mesh (both fields 0) mesh and RHS 1e-10, end states 1e-6 Mesh refinement
Mantle heat transport Theta-dependent top slope of the mixing length tests/test_mlt_calibration.py::test_theta_dependent_top_slope_hand_values_and_limits hand values and limits 1e-15 to 1e-13 relative Solid-state convection
Mantle rheology Arrhenius viscosity, yield stress, NumPy and JAX kernels tests/test_rheology_verification.py (its reference_pinned tests) reference limits, NumPy implementation 1e-12 relative (one bound 1e-9) Solid-state convection
Mantle rheology Saturating integral activation enthalpy limits (\(P_\text{decay} \to \infty\) and \(V_0 = 0\)) tests/test_rheology_verification.py::test_saturating_activation_enthalpy_limits Karato and Wu (1993)13, eq. 1; Armann and Tackley (2012)9, eq. 1; Tackley et al. (2013)20, eqs. 2-3 and Table 2 0 ulp at \(P_\text{decay} = \infty\) and \(V_0 = 0\); \(0 \le H_\text{lin} - H \le V_0 P^2 / (2 P_\text{decay})\) Solid-state convection
Mantle rheology 0-D stagnant-lid cooling trajectory and plastic yield-ceiling lid thinning tests/test_rheology_verification_smoke.py::test_0d_stagnant_lid_reference_and_yield_thinning 0-D stagnant-lid ODE (tests/reference/stagnant_lid_0d.py); Korenaga (2009)14, Table 1 and eq. 29; Solomatov and Moresi (2000)19, Table 6 and eq. 23; Moresi and Solomatov (1998)15, eqs. 25-26; Foley and Bercovici (2014)11, eqs. 26-27 fails derived bounds (pinned to 1e-7 relative; cooling trajectory 6.78 % full-run and 5.51 % pre-clamp vs 5 % on interior temperature \(T_i\), 46.93 % pre-clamp and full-run vs 22 % on lid thickness \(d_\text{lid}\); isotherm thinning ratio 1.53 / 1.74 / 1.91 at fixed heating \(H\) and 1.05 / 1.05 / 1.06 at matched \(T_i\) vs 0-D cube-root targets 0.79 / 0.58 / 0.46) Solid-state convection
Mantle heat transport Stagnant-lid internal-heating scaling exponents (\(\beta, \alpha\)) tests/test_mlt_calibration_smoke.py::test_stagnant_lid_scaling_exponents Solomatov and Moresi (2000)19, sec. 3, eqs. 9-12 and Tables 5-6; Korenaga (2009)14, Table 1, eq. 20 and eqs. 29-30; Reese et al. (2005)18, Table 2 and eq. 20 fails acceptance window (pinned to 1e-7 relative; 12-case design-value fit \(\beta = 0.2819 \pm 0.0134\), \(\alpha = 1.0366 \pm 0.0971\) and realized-state fit \(\beta_\text{real} = 0.1128 \pm 0.0286\), \(\alpha_\text{real} = -0.0786 \pm 0.4398\) vs asymptotic window \([0.294, 0.350]\) and \([1.294, 1.350]\); 7-case convecting diagnostic \(\beta = 0.3112\), \(\alpha = 1.1698\)) Solid-state convection
Mantle numerics NumPy and JAX right-hand sides with solid rheology tests/test_rheology_solver_wiring_smoke.py::test_rhs_numpy_jax_parity_rheology_on NumPy implementation on active-solid abe_solid.toml variant 1e-6 relative (law-only on vs off 4e-3 to 2e-2) Numerics
Mantle numerics Round-off perturbation spread with solid rheology tests/test_rheology_solver_wiring.py::test_perturbed_starts_spread 5 perturbed starts \(S_0(1 + k \cdot 10^{-12})\) over 100 yr \(\Delta T_\text{on}(r) \le 3.5 \times 10^{-5}\text{ K}\) (active solid) and \(\le 2.0 \times 10^{-6}\text{ K}\) (default); stored spreads pinned to \(10^{-9}\) relative Numerics
Mantle numerics Two-pass solid rheology RHS call counts and cost tests/test_rheology_solver_wiring.py::test_rhs_call_counts_and_cost per-RHS call-count budget and off-path wall time \((0, +2, +2, +1, +1, +2)\) calls; JAX ratio \(\le 1.5\) Numerics
Mantle numerics Heated sub-solidus segment that ends at the solidus tests/test_phase_boundary_cap.py::test_rate_mode_ends_a_subsolidus_segment_at_the_solidus 1 yr steps (phase_boundary_cap = 'fixed') at rtol \(10^{-10}\) and a physical atol of \(10^{-10}\) \(1.3 \times 10^{-4}\text{ K}\) over the shared output times (observed \(7.60 \times 10^{-5}\text{ K}\)) Numerics
Mantle numerics Long steps while a cell heats to the solidus tests/test_phase_boundary_cap.py::test_rate_mode_takes_long_steps_while_a_cell_heats_to_the_solidus 1 yr steps (phase_boundary_cap = 'fixed') at rtol \(= 10^{-10}\) and a physical atol of \(10^{-10}\) \(1.3 \times 10^{-4}\text{ K}\) over the shared output times and a step of at least 1000 yr before the solidus (observed \(4.69 \times 10^{-7}\text{ K}\), \(4589\text{ yr}\)) Numerics

The command of a case is pytest <test>; the young-layer case, the surface-boundary cases, the 6 cases of Tabulated core, the surface-lid cases, the phase-transition case, the solid-rheology solver-wiring cases and the 2 heated sub-solidus cases in Numerics, and the saturating-activation, top-slope, 0-D stagnant-lid comparison, and internal-heating scaling cases in Solid-state convection each have a subsection with their reference, tolerance, test id and command, and the other cases are described by the prose of their linked section.

pytest -m reference_pinned runs every case. The core-page tests need the SPIDER P-S tables (ARAGOG_TEST_EOS_DIR, Build tests); the stable-layer and Earth-budget cases take minutes.

Checks on stored runs. These tests read stored runs and do not run the code. They are not in the table and carry no reference_pinned marker. tests/test_surface_lid_v8.py compares the lid after its hand-over with 2 runs without the lid: the step between 200 and 400 nodes is at most 0.75 of the step between 100 and 200 nodes, and the late mean of the surface flux lies within 1 % of the reference on 200 and 400 nodes (Surface lid).

tests/test_surface_lid_v5_rows.py::test_stored_rows_keep_their_values_with_the_lid compares all 36 benchmark rows with and without the lid (Surface lid).

tests/test_surface_lid_base_cap.py::test_lid_base_cap_on_the_mesh_and_the_floor is a sensitivity study with no reference: it reads stored runs of the lid base at the melt temperature on 100, 200 and 400 nodes and with floors of 1e2, 1e3 and 1e4 J/kg.

At 3e5 yr the step from 200 to 400 nodes stays below 1.2 (1.024), at 4e5 yr both mesh steps lie above 1.4, and the floors agree within 4e-2 up to 3e5 yr (Surface lid). pytest tests/test_surface_lid_base_cap.py tests/test_surface_lid_v5_rows.py tests/test_surface_lid_v8.py -m "not slow" --validation-plots writes their plot pairs.

Plots written by the tests

A case with a plot pair shows the tested values against their reference in a light and a dark version. The test writes the pair with the same numbers it asserts, through the validation_plot fixture of proteus-mpl. This command writes the pairs of the table to docs/assets/validation/; the pairs of the checks on stored runs come from the command above:

pytest -m reference_pinned --validation-plots

Without --validation-plots the tests assert and write nothing. The figures in docs/figures/vv/ are made by the scripts in tools/verification/, not by the tests.


  1. D. L. Turcotte, G. Schubert, Geodynamics, 3rd edition, Cambridge University Press, 2014. SciX. ↩↩↩↩

  2. Buono, A.S. & Walker, D., The Fe-rich liquidus in the Fe-FeS system from 1 bar to 10 GPa, Geochimica et Cosmochimica Acta, 75, 2072-2087, 2011. SciX. ↩↩

  3. Chen, B., Li, J. & Hauck, S.A., Non-ideal liquidus curve in the Fe-S system and Mercury's snowing core, Geophysical Research Letters, 35, L07201, 2008. SciX. ↩↩

  4. Fei, Y., Bertka, C.M. & Finger, L.W., High-Pressure Iron-Sulfur Compound, Fe\(_3\)S\(_2\), and Melting Relations in the Fe-FeS System, Science, 275, 1621-1623, 1997. ↩↩

  5. Mashino, I., Miozzi, F., Hirose, K. et al., Melting experiments on the Fe-C binary system up to 255 GPa: Constraints on the carbon content in the Earth's core, Earth and Planetary Science Letters, 515, 135-144, 2019. SciX. ↩↩↩

  6. Mori, Y., Ozawa, H., Hirose, K. et al., Melting experiments on Fe-Fe\(_3\)S system to 254 GPa, Earth and Planetary Science Letters, 464, 135-141, 2017. SciX. ↩↩↩

  7. Nakajima, Y., Takahashi, E., Suzuki, T. et al., "Carbon in the core" revisited, Physics of the Earth and Planetary Interiors, 174, 202-211, 2009. SciX. ↩

  8. Oka, K., Hirose, K., Tagawa, S. et al., Melting in the Fe-FeO system to 204 GPa: Implications for oxygen in Earth's core, American Mineralogist, 104, 1603-1607, 2019. SciX. ↩↩

  9. M. Armann, P. J. Tackley, Simulating the thermochemical magmatic and tectonic evolution of Venus's mantle and lithosphere: Two-dimensional models, Journal of Geophysical Research: Planets, 117, E12003, 2012. SciX. ↩

  10. Christensen, U.R., Holzwarth, V. & Reiners, A., Energy flux determines magnetic field strength of planets and stars, Nature, 457, 167-169, 2009. SciX. ↩

  11. B. J. Foley, D. Bercovici, Scaling laws for convection with temperature-dependent viscosity and grain-damage, Geophysical Journal International, 199, 580–603, 2014. SciX. ↩

  12. Greenwood, S., Davies, C.J. & Mound, J.E., On the evolution of thermally stratified layers at the top of Earth's core, Physics of the Earth and Planetary Interiors, 318, 106763, 2021. SciX. ↩↩↩↩↩↩

  13. S.-I. Karato, P. Wu, Rheology of the Upper Mantle: A Synthesis, Science, 260, 771–778, 1993. SciX. ↩

  14. J. Korenaga, Scaling of stagnant-lid convection with Arrhenius rheology and the effects of mantle melting, Geophysical Journal International, 179, 154–170, 2009. SciX. ↩↩

  15. L. Moresi, V. Solomatov, Mantle convection with a brittle lithosphere: thoughts on the global tectonic styles of the Earth and Venus, Geophysical Journal International, 133, 669–682, 1998. SciX. ↩

  16. Nimmo, F., Energetics of the Core, in Schubert, G. (ed.), Treatise on Geophysics, 2nd edition, Elsevier, 8, ch. 8.02, 27-55, 2015. ↩↩↩↩↩↩↩

  17. Nimmo, F., Thermal and Compositional Evolution of the Core, in Schubert, G. (ed.), Treatise on Geophysics, 2nd edition, Elsevier, 9, ch. 9.08, 201-219, 2015. ↩↩

  18. C. C. Reese, V. S. Solomatov, J. R. Baumgardner, Scaling laws for time-dependent stagnant lid convection in a spherical shell, Physics of the Earth and Planetary Interiors, 149, 361–370, 2005. SciX. ↩

  19. V. S. Solomatov, L.-N. Moresi, Scaling of time-dependent stagnant lid convection: Application to small-scale convection on Earth and other terrestrial planets, Journal of Geophysical Research, 105, 21795–21817, 2000. SciX. ↩↩

  20. P. J. Tackley, M. Ammann, J. P. Brodholt, D. P. Dobson, D. Valencia, Mantle dynamics in super-Earths: Post-perovskite rheology and self-regulation of viscosity, Icarus, 225, 50–61, 2013. SciX. ↩

  21. S. Anzellini, A. Dewaele, M. Mezouar, P. Loubeyre, G. Morard, Melting of Iron at Earth's Inner Core Boundary Based on Fast X-ray Diffraction, Science, 340, 464-466, 2013. SciX. ↩

  22. F. GonzΓ‘lez-Cataldo, B. Militzer, Ab initio determination of iron melting at terapascal pressures and Super-Earths core crystallization, Physical Review Research, 5, 033194, 2023. SciX. ↩↩