ADEPTS: An auto-differentiable framework for time-dependent nonlinear thermo-chemical mantle convection inversion
Abstract
Time-dependent mantle-dynamics inversion must address the high dimensionality of the initial state, nonlinear rheology, and gradient propagation through long-term thermo-mechanical evolution. We develop ADEPTS, a two-dimensional staggered-grid finite-difference framework for mantle-dynamics inversion based on automatic differentiation. The forward model solves incompressible Stokes flow, temperature advection-diffusion, and compositional advection with temperature- and strain-rate-dependent viscosity and plastic yielding. For the nonlinear Stokes system, we compare two gradient strategies: unrolled differentiation through a fixed number of Picard iterations and implicit differentiation of the converged discrete residual equations. Numerical experiments show that unrolled differentiation remains stable even when the nonlinear solve is not fully converged, whereas implicit differentiation requires sufficiently accurate nonlinear solutions; otherwise gradient consistency and optimization convergence deteriorate. With sufficiently converged solves, implicit differentiation recovers accurate gradients and reconstruction quality comparable to unrolled differentiation. Joint thermo-chemical twin experiments show that ADEPTS can simultaneously recover a high-dimensional initial temperature field and low-dimensional physical parameters, including compositional density, reference viscosity, and stress exponent, while fitting final-time temperature, surface horizontal velocity, and surface normal stress. These results demonstrate the feasibility of differentiable time-dependent mantle-dynamics inversion and clarify the different convergence requirements of unrolled and implicit differentiation.
Related
- Structure-dependent failure modes of neural priors in acoustic full-waveform inversion
- Explainable and Uncertainty-Aware Computational Geodetic Inversion: A Review and Framework for Multi-Source Earth Observation
- Evaluating CMAQ Representation of Wildfire PM₂.₅ in California (2008–2018): Regime-Dependent Performance and Nonlinear Error Structure
- The role of financial development on carbon emissions in Sub-Saharan African countries in a nonlinear framework
- Nonlinear thresholds in agricultural landscape fragmentation: A GWRF-SHAP diagnostic framework for differentiated rural land management
BibTeX
@misc{ming2026adepts,
title = {{ADEPTS: An auto-differentiable framework for time-dependent nonlinear thermo-chemical mantle convection inversion}},
author = {Zhiying Ming and Jiashun Hu},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.13482},
eprint = {2609.13482},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.13482},
}
RIS
TY - UNPB
TI - ADEPTS: An auto-differentiable framework for time-dependent nonlinear thermo-chemical mantle convection inversion
AU - Zhiying Ming
AU - Jiashun Hu
PY - 2026
DA - 2026/09/11
PB - arXiv
DO - 10.48550/arxiv.2609.13482
UR - https://arxiv.org/abs/2609.13482
AB - Time-dependent mantle-dynamics inversion must address the high dimensionality of the initial state, nonlinear rheology, and gradient propagation through long-term thermo-mechanical evolution. We develop ADEPTS, a two-dimensional staggered-grid finite-difference framework for mantle-dynamics inversion based on automatic differentiation. The forward model solves incompressible Stokes flow, temperature advection-diffusion, and compositional advection with temperature- and strain-rate-dependent viscosity and plastic yielding. For the nonlinear Stokes system, we compare two gradient strategies: unrolled differentiation through a fixed number of Picard iterations and implicit differentiation of the converged discrete residual equations. Numerical experiments show that unrolled differentiation remains stable even when the nonlinear solve is not fully converged, whereas implicit differentiation requires sufficiently accurate nonlinear solutions; otherwise gradient consistency and optimization convergence deteriorate. With sufficiently converged solves, implicit differentiation recovers accurate gradients and reconstruction quality comparable to unrolled differentiation. Joint thermo-chemical twin experiments show that ADEPTS can simultaneously recover a high-dimensional initial temperature field and low-dimensional physical parameters, including compositional density, reference viscosity, and stress exponent, while fitting final-time temperature, surface horizontal velocity, and surface normal stress. These results demonstrate the feasibility of differentiable time-dependent mantle-dynamics inversion and clarify the different convergence requirements of unrolled and implicit differentiation.
ER -