Geodynamo Theory

And the matter of the Electric Universe Hypothesis

What follows is a copy of a message I sent to a mailing list, in response to the peculiar theory that the magnetic fields of the stars and planets are the result of a rotating, charged body, as opposed to the standard scientific explanation of magnetohydrodynamic dynamo generation.

I have used the HTML pre tag to save myself the trouble of reformatting the message for an HTML environment, so this is a faithful reproduction of the original that I sent on the evening of May 14, 1998. However, I have made all of the URLs "live", which is allowed at least by my version of Netscape.

Relevant material referring to the alleged electric nature of stars as well, can be found in another file.

   I think I have already effectively shown that the "electric
star" part of the "electric universe" model has some rather
severe weaknesses; it violates well known, basic principles of
physics, and some of its assertions appear to be directly
contradicted by observation. So, now I wish to expand the
criticism to the generation of planetary magnetic fields.
The two-pronged approach deals first with issues of fundamental
physics, and then with alleged weakness of mainstream theory.


   It is my understanding that, according to the electric
universe model, the earth's magnetic field is generated solely
by the rotation of the earth, on the assumption that it has
a net charge, and therefore generates a dipole magnetic field.
So, the first task at hand is to decide whether or not, in
light of well established fundamental physics, this proposed
mechanism makes physical sense. So consider the case of a
rotating sphere, with a uniform charge distribution over its
surface; the sphere can be either electrically conducting, or

   The case of the non-conducting sphere is the easier of the
two to address. In this case, since each charge is locked in
place relative to the surface of the sphere, as the sphere
rotates, each charge moves in a circular path, and constitutes
a ring current. It is well known that ring currents generate
dipole magnetic fields (as seen by distant observers), so the
magnetic field of such a sphere will be an integral of the
dipoles for each current loop. Thus, the magnetic field of
the sphere, observed from a distance, will be a dipole field,
or one very similar to a dipole. As is usually the case in
physics, this is easier said than done; those of you with
sufficient daring and bravado are welcome to consult the
ultimate text book on electricity and magnetism, "Classical
Electrodynamics", by J.D. Jackson; reference chapter 5, and
problem 5.7 (page 206).

   The case of the conducting sphere is much harder, and is
not presented even in Jackson (which is proof enough that
this is no ordinary problem). As with the non-conducting
sphere, one expects a basic dipole structure, but here the
mobile charges are subject to coriolis accelerations, and
will also be affected by the magnetic field they generate.
So, the motion of the charges becomes much more complex, and
the observed field will be rather deformed from a basic
dipole nature.

   The case of the conducting sphere appears qualitatively
to offer some solace, since the earth's magnetic field is a
deformed dipole, but there is one remaining fundamental problem.
Observers at rest on the surface of the earth observe, and
map, a global deformed dipole magnetic field. But this is a
violation of the well known principle that magnetic fields
manifest themselves only for charges in motion relative to
the observer. In either of the two cases noted above, an
observer at rest on the surface of the earth will not observe
a dipole field. In the case of the non-conducting sphere, such
an observer will see no magnetic field at all. In the case
of the conducting sphere, only the field resulting from the
coriolis drift of the charge carriers will be observed, but
not the main dipole component.

   The fact that a dipole field is seen by observers at rest
on the earth is very awkward for the proposition that the field
is generated by a rotating charged sphere. This result holds
true for the more general case of any observer who is moving
with the same angular velocity of the earth, such that the
charged sphere is at rest with respect to the observer. This
argument provides a strong fundamental reason for believing that
the proposed mechanism for the generation of the earth's
magnetic field does not, in fact, occur in nature.

                        STANDARD THEORY

   Here some vagueness is required, on the grounds that I really
do not know why standard theory is to be mistrusted. Evidently,
and I emphasize that I can only speculate here, the main
argument brought to bear against standard "dynamo theory" is
that the problem has not yet been solved. This is not much of
a criticism, if indeed it is the one advanced against standard

   The standard theory for the generation of stellar and planetary
magnetic fields is "dynamo theory", also known by the formidable
sounding title "magnetohydrodynamics" (MHD). MHD theory is built
around the fact that a flowing charge-neutral, but electrically
conductive fluid, will generate magnetic fields. MHD is the
theoretical description of how that happens. MHD is highly
technical, and not readily amenable to easy, popular level
explications. All I can do in this forum is provide a qualitative
description of the history of the theory, and references for those
who are more interested in quantitative and technical information.
Standard theory holds that fluid circulation in the earth's fluid
outer core is the source of the earth's internal magnetic field.
This idea is generalized to all of the planets and the sun; the
lack of such circulating fluids is accepted as the reason why
Mars & Venus do not have main magnetic fields, nor do any of the
smaller bodies in the solar system (asteroid magnetic fields are
"remnant fields" frozen into surface magnetic minerals, and
cometary fields are built out of the magnetic field entrained
in the solar wind).

   The idea that planetary magnetic fields might result from a
dynamo process was first put forward by Sir Joseph Larmor in 1919.
He also suggested that sunspot magnetic fields (already known from
early work at Mt. Wilson Observatory) were maintained by current
flow around the sunspot, but the discovery of Cowling's Theorem
in 1934 proved this to be wrong. But dynamo theory became
respectable once again due to the pioneering work of W.M. Elsasser
in 1946 & 1947, and E.C. Bullard in 1949, as well as Hannes Alfven,
Cowling, and others. It was in 1970 that Childress and G.O. Roberts
proved that no general antidynamo theorem existed, a great relief
too those who feared that Cowling's Theorem heralded the discovery
of a more general theorem that could prove all dynamo processes to
be impossible. Led by the efforts of G.E. Backus & E.N. Parker,
over the decades from the 50's through the 70's, dynamo theory
expanded into a potentially powerful explanatory tool for
planetary magnetic fields. Researchers began to model the elusive
reversals in polarity in the earth's magnetic fields in the late
1980's, and continued to make them more realistic until P.H.
Roberts and G.A. Glatzmaier were able to reproduce an entirely
realistic reversal in 1995. My purpose for relating the short
historical narrative is to show that dynamo theory is not some
idle guess work, but rather shows the steady progress towards
solution that is to be expected for any really complex and
difficult problem. The fact that is remains unsolved is, by
itself, no criticism; the steady increase in understanding
clearly illustrated in the literature makes it obvious that
researchers are on the right track.

   I have explained this all in more detail, and provide numerous
references, in one of my articles in the FAQ archive,
"On Creation Science and the Alleged Decay of the Earth's Magnetic
Field" [ ].

                        SOME REFERENCES

   There is very little in the way of general material available
on the WWW on MHD in general. Here is a URL that will allow you to
download a PostScript file (143 kb) of a 24-page mathematical
introduction to MHD theory.
[ ]

   However, the specific application to the earth's magnetic field
(geodynamo theory) is much better represented. P.H. Roberts and
Gary Glatzmaier got a lot of attention in late 1995 and early 1996,
as a result of the ability to achieve a computational model of a
more physically realistic magnetic field polarity reversal than had
ever been done before. The most complete description of their work,
and probably the most readable, is here:
[ ]
A short synopsis of the work is here:
[ ]
Their computational work was carried out at the Pittsburgh Super
Computing Center, which also carries a description here:
[ ]

(From the projects main page
you can also access computational models of turbulent convection
inside the sun and the earth).

   Here is a spiffy animated gif image that shows the time evolution
of the magnetic field at the core-mantle boundary. The first URL is
Johns Hopkins University, the second URL displays exactly the same
animated gif, but is at a German site and might load faster for
European readers.
[ ]
[ ]

   Here is the top page of an introduction to core dynamics and
geodynamo modeling, hosted by the American Geophysical Union, and
written by P.H. Roberts:
[ ]

   Here is a Scientific American "ask the experts" response, written
by Gary Glatzmaier, on reversals of the earth's magnetic field.
[ ]

   Now for old fashioned paper, which is a good thing in this case.
I recommend three books, all unfortunately (or fortunately perhaps)
quite technical in nature, and each of which illustrates a different
aspect of MHD and geodynamo theory. These books are most useful to
the mathematically literate.

   "Foundations of Geomagnetism"
   George Backus, Robert Parker, & Catherine Constable
   Cambridge University Press, 1996
   [ ]
   ISBN 0-521-41006-1 hardback (US $64.95 at the Caltech bookstore)
   This book was primarily the work of Parker & Constable, who built
it out of Backus' lecture notes, as a celebration of Backus' 60th
birthday. Backus is one of the pioneers in geodynamo theory, and
this book is the definitive mathematical work on geodynamo theory.
thoroughly up to date, well organized, and a detailed presentation.
369 pages.

   "The Magnetic Field of the Earth"
   Ronald T. Merrill, Michael W. McElhinny & Philip L. McFadden
   Academic Press, 1996
   [ ]
   Subtitle "Paleomagnetism, the Core, and the Deep Mantle"
   ISBN 0-12-491245-1 hardback

   This is an extensive reworking of the original 1983 edition by
Merrill & McElhinny, "The Earth's Magnetic Field". Includes the
Glatzmaier & Roberts results from 1995, to the extent of using one
of their spiffy graphics for the book cover illustration. A complete
and up to date review of the physics of paleomagnetism, geodynamo,
and the earth's core & mantle. Not as much about the mathematical
justification as Backus et al. give, but a lot more about the
physics of the earth's interior. Also discusses the sun and other
planetary dynamos. 527 pages (compared to 395 in 1st ed.)

   "Reversals of the Earth's Magnetic Field"
   John A. Jacobs
   Cambridge University Press, 1994 (2nd edition)
   [ ]
   ISBN 0-521-45072-1 hardback (US $ 64.95 at the Caltech bookstore)

   This is an extensive reworking of Jacobs' own 1984 first edition,
which was published by Adam Hilger. A brief overview of the general
field, a matter treated in far greater detail by Merrill et al.
Extensive examination of the observational evidence that indicates
that the earth's magnetic field has reversed polarity numerous times.
This is a very important topic - if the earth's magnetic field is to
be explained by the simple expedient of a rotating earth with a net
charge, then polarity reversals are another awkward and difficult to
handle problem. Here is a case where standard theory shows itself
very well by its ability to recreate such reversals with physically
realistic computational models. The book predates the Glatzmaier &
Roberts work, which serves to supplement the book nicely. 346 pages
(compared to 230 in 1st ed.)

                             CLOSING REMARKS

   The primary reason presented, so far as I can tell, for favoring
the "electric universe" model, with its "electric stars" and "electric
planets", is that it is superior to standard theory in its ability to
explain the wide variety of observed electromagnetic phenomena. But
this does not seem to be consistent with what I have been shown. I see
a standard theory that has proven to be resilient, and highly successful.
All across the board, standard theory provides a deep and consistent
understanding of even the most complex processes. Meanwhile, by
comparison, the "electric universe", as it has been presented thus far,
is devoid of both physical and mathematical content; no basis for
physical assertions beyond the level of qualitative analogy are offered,
and no mathematical content is offered at all. Since standard theory
is well endowed along these lines, there seems yet to be little reason
for replacing that theory with the as yet poorly developed ideas of the
"electric universe" hypothesis.

   For those of you who think this is too large a message to archive, it
can also be found on the web at the address below, which includes live
links for all URLs.

Tim Thompson

Back to Tim Thompson's FAQ Page
Back to Tim Thompson's Home Page