How long does a magnetic pole reversal take to complete?

Question

Geomagnetic pole reversal is a "regular" phenomenon on Earth. Although the interval pattern has been irregular at times (sometimes with millions of years of difference between change intervals), Earth has settled in the last 20 million years into a pattern of a pole reversal about every 200,000 - 300,000 years (source: NASA). Based on this pattern and the fact that the most recent pole-switch occurred ~780,000 years ago, the Earth is "overdue" for a pole switch.

My question: How quickly does a pole switch occur?

• Is it instantaneous? Does it take years? Millennia?

• Is the current movement of the pole part of a drawn-out switch, or is it just "normal" pole movement? Again, from NASA:

  The magnetic north pole has been creeping northward – by more than 600 miles (1,100 km) – since the early 19th century, when explorers first located it precisely. It is moving faster now, actually, as scientists estimate the pole is migrating northward about 40 miles per year, as opposed to about 10 miles per year in the early 20th century.

• From some previous research, it seems like it doesn't always make a "clean" switch either, with multiple poles appearing globally -- would this multi-pole extravaganza increase the switching time, or is it also normal?

Answer 1

The entire process appears to take 3,000-4,000 years, according to Valet and Fournier's May 2016 review article "Deciphering records of geomagnetic reversals," which was published in AGU's Reviews of Geophysics.

The complex dynamical structure of reversals is supported by sedimentary and volcanic records. Provided that resolution is adequate, the reversal process seems to incorporate one precursor and one rebound prior to and after the transition. However, the three phases are rarely observed in a single record. This is likely to be caused both by field geometry and recording limitations. The duration of each phase does not exceed 3–4 kyr. Therefore, precursors and rebounds can be seen as part of the whole process, and the expression of a continuum of behavior between secular variation, excursions, and reversals. Another possibility is that the dipole collapse is not large enough or that the nondipole/dipole ratio varies.

Better resolution of the process and the time involved will require at least a 10-fold improvement in paleomagnetic sampling resolution, the authors contend. And one of the challenges is finding suitable lava flows or sediment layers that were created during the transition(s) in question.

As for whether current fluctuations indicate the earth is about to have a reversal: We can't tell, because the recent paleomagnetic record includes similar changes (sometimes regional rather than global) that did not result in reversals, Valet and Fournier write.

The present evolution falls within the normal range of field variability that prevailed for the past 2 Ma.

Answer 2

As per our numerical calculations it takes about ~1000 years to completely flip the dynamo, that being said the current models are not even close to the actual parameters in the earth's core because it would take a humongous time to compute way beyond the capability of current computers. We published a paper back in 2014 where we talk about the role of buoyancy in polarity reversals,

http://ceas.iisc.ernet.in/~bsreeni/gji14.pdf

The critical parameter that supports a stable dipole is helicity. Strong buoyancy offsets helicity in the outer core thus depriving it of a stable dipolar magnetic field. If you want to understand more about the terms I am talking about, here is a link to an excellent review/introductory paper

http://ceas.iisc.ernet.in/~bsreeni/currsci10.pdf