ASTRONOMY:
Galactic Prominences on the Rise
Mark Morris*
The central region of our Galaxy contains a massive
concentration of interstellar molecular gas that has been surveyed many
times. The most often used probe of this gas is emission from the most
abundant molecule that can easily be observed, carbon monoxide. On page
106 of this issue, Fukui et al. (1)
extend earlier surveys of CO to a substantially larger region than had
previously been mapped. The initial results reveal surprising new
structures: giant molecular loops located a few thousand light-years
from the galactic center and extending almost a thousand light-years
above the galactic plane (2).
Interstellar gas in our Galaxy--or in any equilibrated and undisturbed
spiral galaxy--resides in a relatively thin rotating disk. It is
confined to this disk by the gravitational force of the stars, which
are at their densest in the central plane of the galactic disk. The
thickness of the gas disk is determined by the degree to which it has
been stirred by supernovae and galactic shocks (both of which generate
large-scale turbulence) and by the Galaxy's magnetic field, which
prevents the gas from collapsing into a much thinner layer. The
magnetic field in the disk of the Galaxy is predominantly parallel to
the disk, supporting the gas against the largely "vertical"
gravitational field (here, "vertical" means perpendicular to the disk).
In 1966, however, Parker (3)
pointed out that this configuration is not stable: Any vertical
indentation in the magnetic field on a sufficiently large scale will
provide a "pool" into which interstellar gas will sink as it slides
down the tipped field lines in response to gravity (see the figure). As
gas is unloaded from the higher lying sections of the field lines, the
field becomes increasingly buoyant and thus rises there (4).
At the same time, as the pooled section gathers mass, it will sink
toward the galactic midplane, deforming the field more, accelerating
the flow of gas into it. As the resulting gas concentrations become
denser, the gas cools relatively quickly, creating dense molecular
clouds or cloud complexes via the Jeans instability (5).
Because star formation in our Galaxy takes place primarily in molecular
clouds, this combined Parker-Jeans instability might play an important
role in fostering much of the star formation.
Loop formation. Successive stages of the Parker instability occurring above the galactic plane. (Left)
An idealized set of magnetic field lines oriented parallel to the
midplane of a disk-shaped galaxy. The real galactic magnetic field is
subject to considerable distortion about the mean field shown. (Middle) Any vertical perturbation to the field lines begins the instability and causes gas to slide down the tilted field lines. (Right)
As the instability develops, large magnetic loops form above the
galactic plane, and the gas concentrated at the base of the magnetic
loops may form molecular clouds. Unless the tops of the loops are
weighed down by dense gas, they will keep expanding into the halo of
the galaxy.
|
The Parker instability is well
established among theorists and operates in the Sun, playing a role in
the generation of arches, loops, and prominences, and possibly in the
production of sunspots. Yet it has not been definitively observed to be
operating on galactic scales for want of a good diagnostic (
6).
Fukui and his collaborators argue that the giant molecular loops they
observe have resulted from the Parker instability. The radial velocity
field that they infer from the Doppler shifts of the CO emission
indicates that these are indeed loops, rather than shells, and that
there is a flow of material along the loops, as is anticipated by the
theory (
7). Also, the size of the galactic loops is
in the range expected for the Parker instability. This is an exciting
development, raising the possibility of directly confirming this
phenomenon on a galactic scale--a scale larger by 12 orders of
magnitude than the magnetic loops seen on the Sun.
Two challenges are presented by this interpretation, however, both of
which might be met with more observational work. The first is that
there is so far no evidence of a magnetic field in the loops, which
could be resolved by measurement of polarized emission caused by
magnetic alignment of dust grains (8). The galactic center is indeed known to have a strong magnetic field (9),
although it has not yet been measured as far out as the loops are
located. Furthermore, the field within 300 or 400 light-years of the
galactic center is vertical, which would not support the Parker
instability. Thus, the presence of the molecular loops farther out--if
indeed they result from a Parker mechanism--necessitates that the field
orientation shift to the horizontal configuration characteristic of the
rest of the galactic disk somewhere between 400 and 1000 light-years
from the center. Also, if the strength of the galactic magnetic field
at the loops remains high relative to the field farther out in the
disk, then we can perhaps understand why this is the place where they
were discovered: Stronger magnetic loops are more buoyant.
The other challenge is to explain why magnetic loops resulting from the
Parker instability are outlined by molecular gas. Descriptions of the
Parker instability have assumed only a diffuse, low-density atomic gas,
because that is what is typically present in the interstellar medium
far above the galactic plane where the instability manifests itself.
Molecular clouds had previously been thought to form by the
Parker-Jeans instability only at the "footpoints" of the loops, where
they join the gas layer in the galactic plane. Molecular gas may have
been levitated from near the galactic gas layer by the buoyancy of the
magnetic field, but if the gas is flowing down the sides of the loop
along the magnetic field lines, why is there so much gas left high
above the galactic plane? One possible answer to this question is that
the rising portion of the magnetic loop has shocked and compressed the
relatively rarified atomic gas in front of it, leading to rapid cooling
and ultimately to a phase transition from atomic to molecular gas. If
so, then the molecular gas defining the loops is constantly being
replenished.
Deformation of the horizontal magnetic field by the Parker instability
generates vertical magnetic field components out of a horizontal field;
thus, it is interesting to contemplate how the newly discovered loops
might relate to the strong vertical field closer to the galactic
center. The magnetic field in the two vertical legs (or "flux tubes")
of a rising loop should persist with opposite polarity (i.e., opposite
field direction), even after the top of the loop has completely broken
out of the galactic gas layer and expanded into the galactic halo. If
those magnetic legs are then transported toward the center of the
Galaxy by the inexorable inward migration of their footpoints (10),
then all the vertical magnetic flux tubes produced by this process will
pile up there. As they are brought into contact, the flux tubes of
opposite polarity can interact with each other, through a process
called field line reconnection, and deposit energy along their
boundary. This provides a possible explanation for another phenomenon
that radio astronomers have studied for more than 20 years: a
population of vertical, radio-emitting, magnetized filaments in the
inner few hundred light-years of the Galaxy (11).
Because it is these filaments that initially led me and other
astronomers to conclude that there exists a vertical magnetic field at
the galactic center, it seems worthwhile to now reconsider the whole
picture from a broader perspective. The loops reported by Fukui et al. may well give us a handle on this complex process.
References and Notes
- Y. Fukui et al., Science 314, 106 (2006).
- For comparison, our distance from the galactic center is ~25,000 light-years.
- E. N. Parker, Astrophys. J. 145, 811 (1966). [CrossRef]
- The buoyancy of the magnetic loop is
augmented by the pressure of the "gas" of galactic cosmic rays. These
essentially light-speed particles inflate the region encompassed by the
loop.
- B. G. Elmegreen, Astrophys. J. 253, 634 (1982). [CrossRef]
- Some relatively nearby magnetic structures
lying well above the galactic plane have been interpreted as
manifestations of the Parker instability, but these are more readily
explained as compressed, magnetized shells lying at the surfaces of
large, expanding bubbles in the interstellar medium caused by
supernovae and other energy injection processes associated with massive
star formation.
- Without access to information about the
velocity perpendicular to our line of sight, there is an ambiguity
about whether this flow goes up or down, but the expectation from the
Parker instability is that it goes toward the galactic plane.
- G. Novak et al., Astrophys. J. 583, L83 (2003). [CrossRef]
- M. Morris, E. Serabyn, Annu. Rev. Astron. Astrophys. 34, 645 (1996). [CrossRef]
- The footpoints are strongly tied to the
gas in the galactic plane, and they will therefore follow that gas as
it slowly migrates inward toward the galactic center by losing angular
momentum through a variety of processes, one of which is the torque
exerted by these magnetic columns that the gas is dragging along.
- F. Yusef-Zadeh, J. W. Hewitt, W. Cotton, Astrophys. J. Suppl. 155, 421 (2004). [CrossRef]
- I thank F. Coroniti for a thought-provoking discussion.
10.1126/science.1133500
The author is in the Department of Physics and Astronomy, University of California, Los Angeles, CA 90095, USA. E-mail:
morris@astro.ucla.edu
