Tuesday, December 27, 2016

The Problem of Predicting Polarity Reversals of the Earth's Geomagnetic FIeld

No automatic alt text available.
Earth's magnetosphere (right) provides protection from solar eruptions, radiation. But during most of a field reversal that protection disappears.

In previous posts I noted that reversals of Earth's dipole field are critical, given the magnetosphere provides protection from incoming solar radiation - especially in the wake of major solar eruptions such as CMEs (coronal mass ejections).  Typically, the magnetosphere will diminish during a geomagnetic field reversal and even B(E)-> 0, in particular because such reversals include an interval wherein no magnetic field exists. This zero field condition may last from 100 to 1000 years, perhaps more. In such a case, the magnetic field -which normally acts as a protective barrier to the flux of energized particles known as the solar wind- ceases to do that. This means that in the event of major solar flares, the particle flux might attain levels that can be lethal to some life forms.

The problem posed by the solar wind's (or flare) energetic particles is to expose many life forms to much higher radiation levels. To give an indication of the magnitude, when a medium intensity (not even terribly large) solar wind normally "blows" along the magnetosphere boundary an effective MHD (magneto-hydrodynamic) generator is produced. All such "generators" are defined by virtue of the free electrons and protons in the solar wind cutting across the Earth's resident magnetic field.

With an induced voltage of this generator up to 150 kV (150,000 V) a total power of more than 1 million megawatts is produced. That is 1 million joules of energy per second. Entering the Earth's atmosphere unabated such energy flux would pose major issues for any organisms, including humans.

It would therefore seem to be a matter of importance to be able to forecast when the next field reversal will occur. The problem is that the only extant evidence for such reversals is preserved in the geologic record but for which major uncertainties are inherent. Specifically, many field reversals in the past have occurred faster than the measured age resolution of the rocks that record them. If then I only know the age of rocks in a particular strata of the paleomagnetic record to +  10ky (ky = thousands of years)  but a given reversal transpires in 2 ky, then that event can be missed in the record.

To fix ideas, the last reversal occurred some 779,000 years ago but the critical behavior of the transitional field  remains difficult to nail down because of the varying paleomagnetic  record. Given these uncertainties it shouldn't surprise anyone that we still can't predict when the next magnetic field reversal will occur.

Now a new paper by Jean-Pierre Valet and Alexandre Fournier, see e.g.


Supplies answers to many of the questions and issues surrounding field reversals and why they are so difficult to pin down. The authors, who published their work in Reviews of Geophysics (see preceding link) reviewed the major features of reversals and then evaluated the compatibility of these findings with recent numerical modeling results.

The authors found the major source of uncertainty arises from the rapidity of the reversals, as indicated with the example I gave above, i.e. in relation to the age of the rock layer. In addition, the pair found the observed decrease in transitional field intensity can reduce the fidelity of the magnetic record especially for sedimentary rocks. In other words, we can't trust the magnetic record to deliver an accurate picture of what's happening.

As to the inability for forecast reversals in the futures, the authors argue isn't yet advanced sufficiently to determine whether the field strength (B[E]) measured during the past 800 years is an indication of whether the next reversal is imminent.  It could also be merely a signal that falls within the range of long term variability.

One of the reasons that forecasting remains a challenge is due to the limited understanding of the processes that generate the Earth's magnetic field.  The processes are estimated to occur on the scale of several decades to a century or at least an order of magnitude shorter than the time interval over which a typical reversal occurs.

Valet and Fournier argue that future progress in understanding Earth's field reversals depends on a number of further investigations including incorporation of new, more sensitive magnetometers, say capable of scanning magnetization at submillimeter scales. They also recommend including estimates of cosmogenic radionuclide production which might confirm the existence and strength of oscillation prior to reversals.

I'd also suggest looking more closely at the Sun's own magnetic reversals, which occur every 22 years on a full dipole reversal cycle (11 years on a half cycle).  These solar reversals might shed light on our own, and perhaps there is some plasma relationship between the two.

Hopefully, humans will figure out how to do forecasts of reversals before the next major CME arrives.

No comments: