Summary: • A geomagnetic excursion is a significant, short-lived change in Earth's magnetic field, distinct from a full reversal. It involves a dramatic drop in field intensity, often to 0-20 percent of normal, and can include pole orientation shifts up to 45 degrees. These events typically last a few thousand to tens of thousands of years. Unlike reversals, excursions are not always recorded globally, partly due to poor registration in sedimentary records and partly because they may not encompass the entire global field, though exceptions like the Laschamp event exist.
• The frequency of excursions is not well known, as their short duration makes them easy to miss in coarse-resolution records, but they are estimated to be ten times more abundant than reversals. Twelve excursions, including the well-described Laschamp event, are documented in the current Brunhes chron, with others reported from the Matuyama, Gauss, and Gilbert chrons, supported by deep drilling cores from Lake Baikal. Scientific opinion on their cause is divided. The dominant hypothesis suggests they are an inherent instability of the dynamo process, where chaotic liquid metal motion in the core tangles field lines, transferring energy from the main dipole to higher-order multipole configurations, thus weakening the surface field. An alternative hypothesis by David Gubbins posits that excursions involve field reversal only in the liquid outer core, whereas full reversals also affect the inner core.
• A minority view, held by figures like Richard A. Muller, proposes external triggers. These include subducting continental slabs, new mantle plumes, or large impact events disrupting the core's flow. Muller and Donald E. Morris specifically suggest large impacts could cause rapid climate change (a little ice age) and water redistribution, altering Earth's rotation and creating velocity shear that disrupts the dynamo. The effects of an excursion include increased surface radiation due to the weakened field, raising Beryllium 10 and Carbon 14 levels. For modern society, hazards mirror geomagnetic storms, threatening satellites, power supplies, and compass navigation. There is also evidence linking excursions to rapid climatic cooling during ice ages.
Summary: • Reversals of Earth's magnetic field, driven by outer core geodynamo instability, are enigmatic phenomena due to their short duration, which makes high-resolution, datable records scarce. The Matuyama-Brunhes (M-B) reversal is the most studied, but dominant sedimentary records suffer from degraded fidelity, smoothing, and temporal smearing. Lava flows provide superior, well-understood "spot" recordings of field geometry (VGPs) and intensity via thermoremanent magnetization, which can be precisely dated using 40Ar/39Ar methods. This study presents new, high-precision 40Ar/39Ar dates from six volcanic sections recording the M-B transition, using modern multicollector mass spectrometry (IH MCMS) and an astronomically calibrated (FCs) standard, to improve the volcanic record's accuracy and precision.
• The new volcanic data, correlated with high-deposition-rate sediment archives and cosmogenic 10Be proxy records from Antarctic ice, reveal a complex evolution spanning at least 22 thousand years, challenging simpler three-phase models. The data establish at least three distinct periods of geodynamo instability. The first occurred around 795 ka, recorded at Punaruu North (Tahiti). A second, separate instability is recorded in lavas from Chile (Quebrada Turbia) and Guadeloupe at 784 ka. The final phase of the M-B reversal, captured at Haleakala (Maui), La Palma (Los Tilos), and Punaruu South (Tahiti), culminated at 773±2 ka. New paleointensity data confirm the magnetic field was exceptionally weak during these transitional phases.
• Integrating these volcanic "snapshots" with continuous sedimentary and ice core records provides a detailed timeline. The initial ~795 ka event correlates with a significant paleointensity drop and a spike in 10Be (cosmogenic nuclide) production, indicating a dipole collapse. The ~784 ka event, inferred to be the onset phase of the final reversal, correlates with transitional fields in multiple sediment cores. This onset initiated a >10 ka period of highly variable, weak, nondipolar fields, culminating in the final, rapid (<4000 years) polarity switch at 773 ka. The clustering of transitional VGPs further suggests that lower mantle structure exerted repeated control over the weak, nondipolar fields that emerged during this extended instability.
Summary: • A palaeomagnetic investigation of the Matuyama-Brunhes (M-B) reversal, preserved in continuous lacustrine sediments from the Sulmona basin in Central Italy, provides direct, high-resolution evidence for the tempo of the transition. 40Ar/39Ar dating of bracketing tephra layers (SUL2-16 at 781.3 ka and SUL2-22 at 791.9 ka) establishes a mean sediment accumulation rate of about 21 cm per ka, yielding a centennial-scale temporal resolution. The record shows two distinct relative palaeointensity (RPI) minima. The older minimum, centered around 793 ka and lasting ~2.5 ka, is interpreted as the M-B precursor event. This was followed by a ~3.2 ka lag before the onset of the younger RPI minimum, which accompanies the final polarity switch.
• The study's main finding is an extremely rapid 180-degree directional change that occurred during the terminus of the upper RPI minimum. This polarity flip is recorded sharply between 152 cm and 154 cm—a stratigraphic interval of less than 2 cm—with no intermediate virtual geomagnetic poles (VGPs) documented. Based on the calculated sedimentation rate, this full 180-degree transit occurred in "much less than a century," a duration comparable to an average human life. This implies a rate of change exceeding 2 degrees per year, which is an order of magnitude faster than current geodynamo models suggest.
• This study reports the first compelling sedimentary evidence supporting the 'rapid transitional field change' (RTFC) hypothesis, whose validation was previously limited to uncertain data from lava flows. The results demonstrate that even continuous sampling at centennial-scale resolution was insufficient to capture the details of the rapid transit, suggesting that many sedimentary records are too smoothed to resolve the true speed of a reversal. The interpolated age for the M-B boundary in this record is 786.1 ka, an age distinctly older than widely accepted estimates derived from astronomical tuning (773.1 ka) or lavas (776 ka).