Long-term persistency of a strong non-dipole field in the South Atlantic

Earth's magnetic field, generated by a self-sustaining dynamo in the core, is crucial for shielding the planet from harmful solar radiation. While predominantly dipolar, significant non-dipolar features exist, most notably the SAA. This anomaly is characterized by low field intensity and high directional variability, reducing the protective shielding. Understanding the longevity of such features is important for assessing long-term variations in the field and their potential impact on technological systems. Historical, archeomagnetic, and paleomagnetic records offer insights into the temporal evolution of the geomagnetic field, from decades to millions of years. While some non-dipole structures appear persistent in global geomagnetic field models when averaged over time, the long-term stability of the SAA and associated hemispheric asymmetries remains debated. Previous studies have hinted at the recurrence of SAA-like features over the past millennium and across various time intervals (Holocene, 28-48 ka, past 100 ka, and 5.3-23 Ma). These structures at the Earth's surface reflect changes at the core-mantle boundary (CMB) in the form of reversed flux patches (RFPs). However, the persistence of these characteristics on million-year timescales needs further investigation. Some studies suggest hemispheric field asymmetries over the past 10 million years, with evidence from Saint Helena and Tristan da Cunha indicating anomalous behavior in the South Atlantic region, possibly related to CMB thermal heterogeneity influenced by the African large low shear velocity province (LLSVP). This study aims to address the open question of the long-term persistence of the SAA and its associated hemispheric asymmetries by providing high-quality paleomagnetic data from Trindade Island, a young volcanic island in the South Atlantic. This data, in conjunction with existing data, will be used to construct regional PSV and TAF models, further analyzing the relationship between the intensity and position of reversed magnetic flux patches and the PSV latitudinal structure.

Numerous studies have investigated the South Atlantic Anomaly (SAA) and its temporal behavior. Research using archaeomagnetic data from Southern Africa and Brazilian speleothem and sedimentary records suggests SAA recurrence over the past millennium. Paleomagnetic reconstruction models have also revealed SAA-type structures episodically during the Holocene, the 28–48 ka interval, the past 100 ka, and the 5.3–23 Ma interval. These studies highlight the complex nature of the geomagnetic field and the potential for long-term variations in non-dipolar features. The relationship between these surface features and underlying structures at the core-mantle boundary (CMB) has been explored, linking them to the time-dependence of reversed flux patches (RFPs). Several studies have analyzed the paleosecular variation (PSV) and time-averaged field (TAF) on million-year timescales. Analyses of lava flows from Saint Helena Island have revealed anomalous behavior in the South Atlantic, marked by low paleointensity and high dispersion of virtual geomagnetic poles (VGPs). Hypotheses suggest a connection between this longevity of anomalous field variability and thermal heterogeneity at the CMB influenced by the African LLSVP. Dynamo models incorporating lower mantle heterogeneity have partially recovered the locations of persistent geomagnetic surface intensity minima, but the SAA's relatively high latitude remains more challenging to model. Paleosecular variation (PSV) studies, employing paleomagnetic data from igneous rocks, offer valuable insights into the time-averaged field’s temporal evolution and morphology. Analyses of 0–10 Ma datasets allow investigation of longitudinal and latitudinal variations with minimal influence from plate tectonics. However, the global study of PSV and TAF is limited by the paucity of paleomagnetic data, particularly in the Southern Hemisphere, underscoring the need for additional data from this region.

This study expands the Southern Hemisphere paleomagnetic database with high-quality paleodirectional data from Pliocene–Pleistocene volcanic rocks at Trindade Island in the South Atlantic Ocean. The island's geological units range in age from 0.059–3.9 Ma, offering a valuable time window for analysis. Samples were collected from 17 paleomagnetic sites across four geological units. Alternating field (AFD) and thermal demagnetization (THD) techniques were employed to isolate the characteristic remanent magnetization (ChRM) components. Demagnetization diagrams guided the selection of ChRM directions, with a total of 119 specimens yielding acceptable results (MAD ≤ 10°, DANG ≤ 5°). Site-level directions were determined using Fisher statistics, with two sites rejected due to not meeting the selection criteria (k ≥ 50, n ≥ 5) for PSV and TAF analysis. Plate motion corrections were applied using the NNR-MORVEL56 model to calculate VGPs. VGP dispersion (Sg) was calculated with bootstrap confidence limits to detect anomalous directional variability, and compared to the global 0–10 Ma database. Model G, describing the average latitudinal pattern of PSV, was adapted to assess the latitudinal behavior of Sg data. The inclination anomaly (ΔI) was determined, representing the deviation of the mean direction from the expected geocentric axial dipole (GAD) direction. Zonal TAF models were fitted to the ΔI data to analyze the latitudinal field structure. To investigate Atlantic–Pacific hemispheric asymmetries, the analysis was extended globally, separating the data into Atlantic and Pacific hemispheres. Model G parameters and zonal TAF models were determined separately for each hemisphere, and the axial dipole dominance (AD/NAD) ratio was estimated for both. Finally, idealized geomagnetic field scenarios were created by superimposing equatorially anti-symmetric flux patches onto a background axial dipole field. These models helped evaluate the influence of the location and intensity of geomagnetic flux patches on the latitudinal behavior of VGP dispersion.

The paleomagnetic data from Trindade Island revealed a VGP dispersion significantly higher than that observed in other PSV studies at similar latitudes but outside the SAA region. This high dispersion, similar to that found on Saint Helena Island, suggests an anomalous PSV behavior in the South Atlantic region that has persisted over at least the past 10 Myr. The inclination anomaly (ΔI) from Trindade Island was consistent with a GAD field model and the global 0–10 Ma TAF model. However, comparisons with other studies at similar longitudes showed differences in ΔI values, suggesting regional variations in the long-term geomagnetic field. Analysis of the global 0–10 Ma database revealed Atlantic–Pacific hemispheric asymmetries in both PSV and TAF. The Atlantic hemisphere exhibited higher VGP dispersion and a greater number of negative inclination anomalies than the Pacific. The AD/NAD median ratio was lower for the Atlantic hemisphere compared to the Pacific, indicating weaker average dipole dominance in the Atlantic and supporting the presence of strong NAD effects. Synthetic geomagnetic field models, which considered variations in the location and intensity of geomagnetic flux patches, showed that the latitudinal pattern of VGP dispersion is greatly affected by the positions and amplitudes of these flux patches at the core-mantle boundary (CMB). Comparisons of the results from the models to real-world data suggest that the geomagnetic field in the last 10 Myr has been clearly influenced by normal flux patches (NFPs) and/or reversed flux patches (RFPs) at the CMB.

The findings support the hypothesis of a long-lived non-dipole field in the South Atlantic, potentially linked to heterogeneous boundary conditions at both the top (lowermost mantle) and bottom (inner core) of the outer core. The enhanced geomagnetic variability observed in the South Atlantic region over the past 10 Myr could be attributed to core fluid upwelling under the African LLSVP, resulting in the formation of pairs of RFPs and NFPs at the CMB. The observed Atlantic–Pacific hemispheric asymmetries in PSV and TAF suggest a stronger bottom-up control on geodynamo temporal variability, possibly driven by the inner core's influence on outer core dynamics. However, mantle heterogeneities likely play a significant role in shaping the long-term morphology of the geomagnetic field. The synthetic model results indicate that the observed VGP dispersion patterns could be explained by the recurrent presence of intense low-latitude flux patches at the CMB, which are not necessarily linked to consistently weak surface fields. The findings of the study highlight the complex interplay between core and mantle processes in shaping the long-term behavior of the Earth’s magnetic field.

This study provides compelling evidence for the long-term persistence of anomalous geomagnetic field behavior in the South Atlantic region over the past 10 Myr. The high VGP dispersion and Atlantic–Pacific hemispheric asymmetries suggest a significant influence of both inner core dynamics and lower mantle heterogeneities on the geodynamo. The results support the hypothesis of a stronger bottom-up control on the geodynamo's temporal variability. Future research should focus on further refining the synthetic models and integrating additional paleomagnetic data from the Southern Hemisphere to improve the understanding of the complex interactions between the Earth's core and mantle.

The relatively small number of paleomagnetic sites from Trindade Island might influence the calculated VGP dispersion. The synthetic geomagnetic field models, while useful for exploring the effects of flux patches, are simplifications of the complex processes occurring within the Earth's core. The limited data coverage in certain regions, such as the high-southern latitudes, restricts the spatial resolution of the average field structure in those areas.