These events were noted by Courtillot and others in 1978. The clearest ones, observed all over the world, happened in 1969, 1978, 1991, and 1999. Data before 1969 is scarcer, but there is evidence of other global jerks in 1901, 1913, and 1925. Other events in 1932, 1949, 1958, 1986, and 2003 were detected only in some parts of the world. These events are believed to originate in the interior of the Earth (rather than being due to external phenomena such as the solar wind); but their precise cause is still a matter of research.
The name "jerk" was borrowed from dynamics, where it means the rate of change of the acceleration of a body, that is, the third derivative of its position with respect to time (the acceleration being the second derivative); or, more specifically, a sudden and momentary spike (or dip) in that rate.
Jerks seem to occur in irregular intervals, on average about once every 10 years. In the period between jerks, each component of the field at a specific location changes with time t approximately as a fixed polynomial of the second degree, A t2 + B t + C. Each jerk is a relatively sudden change (spread over a period of a few months to a couple of years) in the A coefficient of this formula, which determines the second derivative; and usually in B and C coefficients as well.
The strength of each jerk varies from location to location, and some jerks are observed only in some regions. For example, the 1949 jerk was clearly observed at Tucson (North America, long. 110.93°), but not at Chambon la Forêt (Europe, long. 2.27°). Moreover, the global jerks seem to occur at slightly different times in different regions; often earlier in the Northern hemisphere than in the Southern hemisphere.
These events are believed to be caused by changes in the flow patterns of the liquid outer core of the Earth.  Another theory is that they are due to torsional oscillations in the solid inner core of the Earth. There have been claims that they are connected to strong earthquakes.
- P. De Michelis, R. Tozzi, and A. Meloni (2005), Geomagnetic jerks: observation and theoretical modeling. Memorie della Società Astronomica Italiana, Vol. 76, pp. 957–960. Online version accessed on 2009-07-10.
- Mioara Mandea, Eric Bellanger and Jean-Louis Le Mouël (2000), A geomagnetic jerk for the end of the 20th century? Earth and Planetary Science Letters, Volume 183, Issues 3-4, Pages 369–373. doi:10.1016/S0012-821X(00)00284-3
- Nils Olsen and Mioara Mandea (2007), Investigation of a secular variation impulse using satellite data: The 2003 geomagnetic jerk. Earth and Planetary Science Letters, Volume 255, Issues 1-2 Pages 94–105. doi:10.1016/j.epsl.2006.12.008
- Bloxham, J., Zatman, S., & Dumberry, M. 2002, Nature, 420, 65
- Fabio Florindo, Paola De Michelis, Antonio Piersanti, Enzo Boschi (2005), Could the Mw= 9.3 Sumatra Earthquake Trigger a Geomagnetic Jerk? EOS, Transactions of the American Geophysical Union, volume 86, issue 12, doi:10.1029/2005EO120004