Newsgroups: sci.physics.plasma
From: fcrary@benji.Colorado.EDU (Frank Crary)
Organization: University of Colorado, Boulder
Subject: Re: Can Gravity be Induced?

In article <39u8d2$ke6@mojo.eng.umd.edu>,
Christoph Keller <keller@noao.edu> wrote:
>> the average fusion rate. So it is entirely possible
>> that fusion in the sun's core varies with time, on
>> a scale of decades. If we happened to be looking
>> during a time of minimal fusion, we would see
>> far fewer neutrinos than we would expect, based on
>> the observed, average energy output. Nor would
>> this imply periodic variations in solar energy
>> output: The process of heat conduction would
>> average out these decade-long variations reached
>> the observable regions of the sun.

>This explanation is very unlikely by now thanks to accurate
>helioseismological observations. Global oscillations do not have the
>time lag of temperature fluctuations.

I'm not so sure how convincing this is. How deep to the global
modes penetrate? It isn't really my field, but my understanding
was that even the global modes don't get all the way to the
core (although they obviously must go fairly far down: I think
you can get global oscillations even if the waves don't
propagate beyond the top 10% or so of the sun, but in that
case they would be rather weak compared to higher-order
modes.) If we are talking about periodic variations
limited to the innermost regions, would it show up in
the global modes? (I.e. do the global modes really
sound the entire sun?) Of course, that theory would not
be without flaws: It would require some mechanism for
keeping the oscillations localized in the core. Otherwise
they would drive global modes closer to the surface. Possibly,
the edge of the convective region might provide a boundary
to localize core oscillations, but I can't think of
anything else. So another important question is
whether the global modes sound the sun below the
convective region.

>...Furthermore a low-temperature
>core model would need a helium abundance that is lower than the
>primordial helium abundance (at least the more popular non-standard
>solar models).

I can see this for a core which is low-temperature on average.
But why would a different helium abundance be required
for a core which was, on average, the expected temperature
but which had periodic variations?

>> As well as the non-thermal heating processes, it is also possible
>> that the corona is simply not as hot as it looks. This is
>> actually a good exercise for students of plasma physics.
>> "Temperature" is really the average kinetic energy of the
>> particles. But often, the average doesn't mean much.
>> ...Now observations of the corona's temperature
>> are based on a particular sort of weighted average
>> (weighted by collision cross-sections and the emissions
>> they excite) while heat transfer equations are based
>> on a different, weighted average (based on the
>> kinetic energy transferred by collisions).
>> ...the [two] sorts of averages might be very different:
>> "Temperature", for the purposes of observed, emitted
>> light might be quite different from "temperature"
>> for the purposes of heat conduction.

>I am aware of coronal heating theories that use non-Maxwellian
>distributions (e.g. Scudder 1992, ApJ 398, 319). However, there is
>currently no viable theory on how you get the non-Maxwellian distribution
>at the base of the corona...

Has anyone looked at the inverse problem? I.e. estimating how
what sort of distribution would be required to reconcile
the observed temperatures with thermal heat conduction?
Of course, such a study wouldn't give a unique answer: It
would require assuming a particular sort of non-maxwellian
distribution and estimating the free parameters, but
there is no easy to say what sort of non-maxwellian
would be appropriate. But all the same, it might suggest
the magnitude of the problem. A slightly non-maxwellian
distribution like a kappa function might be easy to
explain in the corona, say from wave-particle
interactions. (Since wave heating has been proposed as
a coronal heating mechanism, I don't think assuming the
presence of a lot of plasma waves would be so radical
an assumption, although the presence of waves of the
appropriate phase velocity might be.)

>...Also note that thermal X-rays from the corona are
>observed, not just emission lines.

That really would be a restriction on non-maxwellian
distributions, since it is sensitive only to the
high-energy particles rather than the bulk of the
plasma (which probably would be maxwellian). But
would a hybrid theory perhaps work? For example,
a slightly non-maxwellian distribution to
reduce but not eliminate the required non-thermal
heat input combined with one of the many, proposed
non-thermal mechanisms? The proposed non-thermal
mechanisms don't add enough heat to explain matters
if you assume a maxwellian plasma, but if a
non-maxwellian plasma could reasonably imply
a factor of (say) two less heat input, would these
ideas become viable?

>>>Rotation cannot produce magnetic fields.

>> While I agree with many of your earlier remarks, this is not
>> correct. The essence of a dynamo is the creation of
>> a magnetic field drawing the required energy from
>> rotational motion. Even in the absence of an initial
>> magnetic field, a dynamo is unstable and would produce
>> a strong magnetic field from even the most trivial
>> disturbance.

>I agree that in a very general term you may create magnetic fields
>from certain plasmas. We were talking about the solar case, and I am
>not aware of any theory that may create magnetic fields from rotation
>and does that in a period that is short with respect to the age of the sun.
>(please note: I am aware of theories on the generation of seed fields.)
>The usual dynamo equations derived in magnetohydrodynamics do not
>contain a term that creates magnetic fields from a field-free state.
>I would be interested to hear how you get a term that creates fields.

I think the real catch is "from a field-free state". Clearly,
if there was never a magnetic field at all, no configuration
of plasma circulation could produce one. I was thinking of
situations where a pre-existing "seed" field existed and
where the system would support growth of the seed field.
If I'm remembering correctly, such an unstable configuration
could start with a trivial magnetic field and the
instability would grow until the field was quite strong.
Unless I'm misremembering something, this would be all
that was required, since there certainly was some
sort of seed field. On the other hand, this may be
an academic consideration: Observations of the
interstellar medium, specifically of magnetic fields
in star-forming regions, suggest that we aren't
taking about a trivial, initial magnetic field, but
rather a fairly strong one. Since it is entirely
possible for a solar dynamo to sustain or slightly
increase such an initial field, perhaps the
issue of a large magnetic field evolving from a
small one is irrelevant. On the other hand, the
bulk of the sun's magnetic field seems to
be very time-dependent: The dipole term flipping sign
back and forth between solar cycles, and the higher-order
terms growing and decaying (apparently) on the same
time-scale but out of phase. So possibly, the
issue of a growing magnetic field from on unstable
configuration might be significant. On the other hand,
it might just be a very complex version of the as-yet
unexplained polarity flips also observed in planetary
magnetic fields: Not the creation of magnetic
fields but their alteration.

Frank Crary
CU Boulder