Using paleomagnetic observations and geodynamo simulations to assess Earth's magnetic field morphology and variability

Hannah Rogers ⁽¹⁾, Christopher Davies ⁽¹⁾, Catherine Constable ⁽²⁾, Stephen Mason ⁽¹⁾, Thomas Frasson ⁽¹⁾

Using paleomagnetic observations and geodynamo simulations to assess Earth's magnetic field morphology and variability

Hannah Rogers ⁽¹⁾, Christopher Davies ⁽¹⁾, Catherine Constable ⁽²⁾, Stephen Mason ⁽¹⁾, Thomas Frasson ⁽¹⁾

(1) University of Leeds, School of Earth and Environment

(2) University of California San Diego

Abstract

Multiple strategies have been suggested for quantitatively comparing numerical dynamo simulations to geomagnetic field models and paleomagnetic observations. Observationally-constrained metrics are designed to infer properties of simulations that are required to produce Earth-like behaviour and enable inferences on otherwise inaccessible properties of the geomagnetic field, such as its long-term spatio-temporal behaviour at the core-mantle boundary. However, these criteria are derived from data spanning differing timescales with fundamentally different spatio-temporal resolution, are often applied in isolation, and may not be independent assessments, so that holistic syntheses of simulated and observed field variations are currently lacking. In this work, we apply 14 existing criteria measuring field morphological and variability properties on centennial to million-year timescales to a database of 207 dynamos. Individual metrics are matched over various ranges of core-mantle boundary dipolarity (𝑓𝑑 ) and its temporal variability (𝛿 𝑓𝑑 ), though no single range conforms with all proposed metrics. The greatest overlap between simulations matching disparate criteria occurs for 𝑓𝑑 = 0.50−0.64, which is lower than the modern field value of 0.65−0.71 (1900-2025). Earth-like dynamos tend to have magnetic/kinetic energy ratio >1, consistent with a MAC force balance, and require a magnetic Reynolds number 𝑅𝑚 = 750−1200 to match the secular variation timescale. Simulations in our dataset exhibit reduced temporal variability at moderate dipolarity (𝑓𝑑 ∼ 0.5) compared to inferences from global field models, which hinders their capacity to produce Earth-like polarity reversals. This is a correction of the original manuscript where the 𝑓𝑑 value in 2025.0 was incorrectly stated. We thank Julien Aubert for his careful reading and suggestion which is now incorporated.

Keywords

• Geodynamo simulations
• Geomagnetic field models
• PaleoSecular Variation

Preview

Loading PDF preview...