Thermal state reconstruction--reversing convection to recover the thermal structure of the mantle at an earlier geologic time--is an important tool to understand the evolution of mantle convection and its relation to seismic tomographic images and observations at the surface. Thermal state reconstructions are computationally expensive. Here we transformed the basic computational element, numerical solvers, into neural operators, a class of machine learning models for learning mappings between function spaces. Focusing on a specific architecture, Fourier Neural Operators, we demonstrate that they can represent not only a surrogate model like the Stokes system of equations using a purely physics informed approach, but also discover operators without explicit mathematical formulations or even ill-posedness from data, including the direct mapping between two convecting thermal states separated by a long time interval much larger than the Courant-Friedrichs-Lewy condition and its reversal. These neural operators significantly accelerate forward and inverse convection modelling by transforming forward physical processes into surrogate models with lower complexity while utilizing auto-differentiation to calculate gradients. With this framework, we demonstrate the strengths and weaknesses of four methods for thermal state reconstructions: reverse buoyancy, reverse convection operator, an inversion with only the terminal thermal state, and a joint inversion with the terminal thermal state and surface velocity evolution. The reverse convection operator is shown to perform poorly in the presence of observational noise, but the joint inversion overcomes this limitation. The joint technique could probably become a solution to large-scale thermal state inversion problems using seismic tomography and plate tectonic reconstructions.
ADEPTS: An auto-differentiable framework for time-dependent nonlinear thermo-chemical mantle convection inversion
Southern University of Science and Technology(南方科技大学);Guangdong Provincial Key Laboratory of Geophysical High-resolution Imaging Technology, Southern University of Science and Technology(南方科技大学地球物理高分辨率成像技术广东省重点实验室)
Mantle Convection and Nightside Volcanism on Lava World K2-141 b
University of Oxford(牛津大学);Bayerisches Geoinstitut, University of Bayreuth(拜罗伊特大学巴伐利亚地球研究所);Kapteyn Astronomical Institute, University of Groningen(格罗宁根大学卡普坦天文研究所);Institute of Geophysics, Department of Earth and Planetary Sciences, ETH Zurich(苏黎世联邦理工学院地球与行星科学系地球物理研究所);School of Physics and Astronomy, University of Leeds(利兹大学物理与天文学学院)
Morphology, interactions, and evolution of laminar thermal plumes revealed by time-resolved volumetric velocity measurements and analysis: Application to Earth's mantle
Department of Earth, Planetary, and Space Sciences, University of California, Los Angeles(加州大学洛杉矶分校地球、行星与空间科学系);Department of Earth and Planetary Sciences, Harvard University(哈佛大学地球与行星科学系)
Iron spin crossover in ferropericlase and its effect on lower-mantle thermal conductivity
Carnegie Institution for Science(卡内基科学机构);Florida International University(佛罗里达国际大学);University of Münster(明斯特大学);The University of Texas at Austin(德克萨斯大学奥斯汀分校);The University of Edinburgh(爱丁堡大学);Deutsches Elektronen-Synchrotron (DESY)(德国电子同步辐射研究所);Albert-Ludwigs Universität Freiburg(弗赖堡大学);European XFEL(欧洲X射线自由电子激光设施);Univ. Lille, CNRS, INRAE, Centrale Lille(里尔大学);Sorbonne Université, Muséum National d’Histoire Naturelle(索邦大学,法国国家自然历史博物馆);Helmholtz-Zentrum Dresden-Rossendorf(德累斯顿-罗斯多夫亥姆霍兹中心);Université Grenoble Alpes(格勒诺布尔阿尔卑斯大学)
Modelling of thermal stratification at the top of a planetary core: Application to the cores of Earth and Mercury and the thermal coupling with their mantles