Geomagnetic Jerks: When Earth's Magnetic Field Suddenly Changes Course

Earth's magnetic field is not a fixed bar magnet. It drifts, strengthens and weakens all the time, driven by molten iron flowing in the outer core roughly 3,000 km below our feet. Most of that change is slow and steady. But every few years the trend itself suddenly bends. Geophysicists call these moments geomagnetic jerks. Here's what they are, what's known, and why they come up in Earth-change conversations.

What a jerk actually is

Think of it like driving. The field's strength and direction are the car's position. How fast it changes year to year (the secular variation) is the speed. The change in that speed is the secular acceleration. A jerk is an abrupt change in that acceleration: the British Geological Survey defines jerks as "abrupt changes in the second time derivative, or secular acceleration, of the magnetic field that arises from sources inside the Earth" (Macmillan, BGS). On an observatory chart of secular variation, a jerk shows up as a sharp V or upside-down V where two straight-line trends meet (Torta et al. 2015).

The name is borrowed from mechanics, where the rate of change of acceleration is literally called "jerk" (Torta et al. 2015). A jerk does not mean the field suddenly lurches in strength. The changes are subtle, and you need years of careful measurements to see them.

How they're detected

Ground observatories. The classic method is to take monthly or annual averages from magnetic observatories, which smooths out the noise from the Sun and the upper atmosphere, then look for that V-shaped kink in the trend (Torta et al. 2015). The 2014 jerk was first spotted this way at Spain's Ebro observatory while staff were preparing routine data files (Torta et al. 2015).

Satellites. Since 1999 the field has been monitored continuously from space, most recently by ESA's Swarm mission (Livermore et al. 2020). Researchers turn Swarm data into "virtual observatories" at 440 km altitude, which fill in the huge gaps over oceans where there are no ground stations (Pavón-Carrasco et al. 2021).

A short history

• 1969: the first recognized jerk, reported by Courtillot and colleagues in 1978 from European observatory data (Macmillan, BGS).
• 1978 and 1991: detected globally; a 1999 event was seen only locally (Brown et al., G-cubed 2011).
• Earlier and later: jerks are also reported around 1925 and possibly 1901 and 1913 (Macmillan, BGS), plus 2003, 2007 and 2011 (Torta et al. 2015).
• 2014: announced by Torta and colleagues, centered around 2014.0 (Torta et al. 2015).
• 2017: a regional jerk, clearest in the Pacific (Pavón-Carrasco et al. 2021).
• 2019–2020: a more global jerk, seen in both ground and Swarm data. The authors note predictions had suggested a new jerk roughly every 3–4 years (Pavón-Carrasco et al. 2021).

The standard scientific review of the topic is Mandea et al. 2010, Space Science Reviews.

What causes them

For decades nobody could reproduce jerks in a model. In 2019, Julien Aubert and Christopher Finlay published simulations in Nature Geoscience that did (Aubert & Finlay 2019). In their model, sudden releases of buoyancy deep in the core send out packets of Alfvén waves, a kind of wave that travels along magnetic field lines in a conducting fluid. When these waves reach the top of the core, they focus toward the equator and along lines of strong magnetic flux, causing sharp changes in core flow that we see at the surface as jerks. The simulation took the equivalent of 4 million hours of computing (CNRS). Other researchers have proposed related rapid core waves, including magneto-Coriolis and torsional waves; which ones dominate is still debated (Mandea et al. 2010).

The pole on the move

The north magnetic pole was first located in the Canadian Arctic in 1831. Between 1990 and 2005 its drift sped up from a historic 0–15 km a year to 50–60 km a year, heading toward Siberia (Livermore et al. 2020). Livermore, Finlay and Bayliff traced this to a tug of war between two patches of magnetic flux at the core surface, under Canada and Siberia. The Canadian patch stretched and weakened, likely because core flow changed between 1970 and 1999 (Livermore et al. 2020). The pace has since eased: NOAA reports the north dip pole averaged about 36 km a year during 2025, close to the World Magnetic Model forecast (NOAA NCEI, Dec 2025).

The same rapid core-flow changes behind jerks are part of what makes the field hard to forecast. In 2019 the World Magnetic Model, used in navigation and phones, needed an unusual early update (Livermore et al. 2020; NOAA NCEI).

The South Atlantic Anomaly

Over South America and the South Atlantic lies the weakest part of the field, the South Atlantic Anomaly. ESA reports that the global field has lost about 9% of its strength over 200 years, and that the anomaly's minimum dropped from about 24,000 to 22,000 nanoteslas between 1970 and 2020 while it grew and moved west (ESA Swarm). NOAA says it deepened by about 30 nT more and shifted about 20 km west in 2025 (NOAA NCEI, Dec 2025).

Is it linked to jerks? One 2020 study found that dips in the anomaly's growth rate lined up with jerks over the past two decades, suggesting the same core processes are involved (Frontiers in Earth Science, 2020). Recent jerks also seem to start as acceleration pulses under West Africa, the South Atlantic and the southwest Indian Ocean (Torta et al. 2015).

Jerks and the Earth-change debate

In the community I follow, a weakening field and a fast-moving pole are read as warning signs. The Ethical Skeptic's ECDO hypothesis (see my explainer, ECDO Explained) treats changes in the core and the magnetic field as possible signals that the core-mantle system is building toward an abrupt event. Others in the space ask whether field weakening hints at an approaching reversal or excursion, like the Laschamp event about 41,000 years ago.

Mainstream geophysics reads the same data differently: jerks are seen as a normal, recurring feature of a working dynamo, recorded every few years since observatories began. ESA says the current weakening does not mean a reversal is imminent (ESA Swarm). Both camps agree on the measurements. They disagree on what they mean and what comes next.

What about the body?

Some light-health voices suggest that shifts in the geomagnetic field affect circadian rhythms, mood or mitochondria. Many animals clearly sense the field for navigation. But I haven't found solid evidence that geomagnetic jerks, which are tiny and slow, have any measurable effect on human health, so treat those claims as a Decentralized View.

Sources

• Aubert, J. & Finlay, C. C. (2019). Geomagnetic jerks and rapid hydromagnetic waves focusing at Earth's core surface. Nature Geoscience 12, 393–398.
• CNRS (2019). Geomagnetic jerks finally reproduced and explained.
• Mandea, M. et al. (2010). Geomagnetic Jerks: Rapid Core Field Variations and Core Dynamics. Space Science Reviews 155.
• Macmillan, S. Geomagnetic Jerks. British Geological Survey.
• Brown, W. et al. (2011). Measurements and uncertainties of the occurrence time of the 1969, 1978, 1991 and 1999 geomagnetic jerks. G-cubed.
• Torta, J. M. et al. (2015). Evidence for a new geomagnetic jerk in 2014. Geophysical Research Letters.
• Pavón-Carrasco, F. J. et al. (2021). Signs of a new geomagnetic jerk between 2019 and 2020 from Swarm and observatory data. Earth, Planets and Space.
• South Atlantic Anomaly areal extent as a possible indicator of geomagnetic jerks in the satellite era (2020). Frontiers in Earth Science.
• Livermore, P. W., Finlay, C. C. & Bayliff, M. (2020). Recent north magnetic pole acceleration towards Siberia caused by flux lobe elongation. Nature Geoscience 13.
• NOAA NCEI (Dec 2025). State of the Geomagnetic Field, World Magnetic Model.
• ESA (2020). Swarm probes weakening of Earth's magnetic field.

← All articles