[Paper Review] Spatial And Temporal Changes Of The Geomagnetic Field: Insights From Forward And Inverse Core Field Models

Nicolas Gillet|arXiv (Cornell University)|Feb 21, 2019

Geomagnetism and Paleomagnetism Studies123 references4 citations

TL;DR

This paper advances understanding of Earth's core dynamics by integrating forward and inverse modeling of the geomagnetic field across interannual to millennial timescales. It demonstrates that data assimilation combining geodynamo simulations with observational data—especially from satellites and archeomagnetic records—can improve estimates of core magnetic field evolution, though challenges remain in resolving short-timescale changes due to external field interference and model limitations in capturing energy cascades and subgrid-scale processes.

ABSTRACT

Observational constraints on geomagnetic field changes from interannual to millenial periods are reviewed, and the current resolution of field models (covering archeological to satellite eras) is discussed. With the perspective of data assimilation, emphasis is put on uncertainties entaching Gauss coefficients, and on the statistical properties of ground-based records. These latter potentially call for leaving behind the notion of geomagnetic jerks. The accuracy at which we recover interannual changes also requires considering with caution the apparent periodicity seen in the secular acceleration from satellite data. I then address the interpretation of recorded magnetic fluctuations in terms of core dynamics, highlighting the need for models that allow (or pre-suppose) a magnetic energy orders of magnitudes larger than the kinetic energy at large length-scales, a target for future numerical simulations of the geodynamo. I finally recall the first attempts at implementing geomagnetic data assimilation algorithms.

Motivation & Objective

• To assess the current resolution and accuracy of geomagnetic field models spanning from archaeological to satellite-era observations.
• To address the challenge of separating internal core signals from external (ionospheric/magnetospheric) sources that obscure short-timescale field changes.
• To evaluate the potential of data assimilation in merging geodynamo simulations with observational data to infer core state and dynamics.
• To identify key limitations in current geodynamo simulations, particularly in resolving energy cascades and subgrid-scale processes.
• To explore the feasibility of using reduced models and stochastic parameterizations to improve data assimilation in underdetermined inverse problems.

Proposed method

• Utilizes forward and inverse modeling of the geomagnetic field using Gauss coefficients derived from satellite (Oersted, CHAMP, Swarm), observatory, archeomagnetic, and sediment records.
• Applies data assimilation techniques—such as ensemble Kalman filters (EnKF), optimal interpolation (OI), and augmented state Kalman filters—incorporating geodynamo simulations and observational constraints.
• Employs reduced models based on Magneto-Coriolis (MC) and Magneto-Archimedes-Coriolis (MAC) waves to analyze core field variations at interannual to centennial timescales.
• Integrates dynamical constraints from geodynamo simulations into field modeling, including the use of a 'dynamo norm' to regularize archeomagnetic field models.
• Incorporates uncertainty quantification for measurements and unmodeled processes, especially in archeomagnetic and sediment records.
• Tests assimilation strategies using multivariate inference and cross-covariances between geodynamo states and observed secular variation (SV) to improve 5-year SV predictions.

Experimental results

Research questions

• RQ1 How accurately can we resolve interannual to millennial changes in the geomagnetic field using current observational and modeling frameworks?
• RQ2 To what extent do external field signals (e.g., ionospheric currents) limit the recovery of rapid, short-wavelength internal field changes in satellite data?
• RQ3 Can data assimilation of geodynamo simulations with observational data improve the estimation of core magnetic field evolution and secular variation?
• RQ4 What are the limitations of current geodynamo simulations in reproducing Earth-like core dynamics, particularly in terms of energy cascade and reversal behavior?
• RQ5 How can reduced models and stochastic parameterizations enhance data assimilation in core dynamics, given the underdetermined nature of the inverse problem?

Key findings

• Satellite data from 1999 onward (Oersted, CHAMP, Swarm) provide near-global coverage but suffer from polar gaps and orbital drift, limiting resolution of long-term external field variations.
• Ground-based and archeomagnetic records exhibit spatial biases—especially toward the Northern Hemisphere and Middle East—leading to sensitivity kernels that are strongly non-uniform.
• Data assimilation using geodynamo simulations can produce 5-year secular variation predictions with improved accuracy, as demonstrated by multivariate inference and Kalman filtering approaches.
• Current geodynamo simulations fail to simultaneously reproduce all key timescales of core dynamics, particularly due to excessive diffusive processes and inability to reach the critical Alfvén number (A < O(1)) required for realistic energy cascades.
• Subgrid-scale parameterizations are essential to reduce the Alfvén number in simulations, but no such models currently exist for turbulent, rotating magnetohydrodynamics with reverse energy cascades.
• Reduced models based on MC/MAC waves are effective tools for analyzing core field changes at interannual to centennial timescales and are well-suited for integration into data assimilation frameworks.