Received 24th May 1998
Abstract.
We present a multi-wavelength study of the Violent Interstellar Medium of
the nearby dwarf galaxy IC 2574, a member of the M81 group of galaxies.
In particular, we concentrate on the most prominent supergiant shell
in IC 2574 which was detected in neutral hydrogen
(H
I) observations obtained with the
Very Large Array (VLA).
This shell is thought to be produced by the combined effects of stellar winds
and supernova explosions.
Massive star forming regions, as traced by H
alpha emission,
are situated predominantly on the rim of this H
I shell.
This supports the view that the accumulated H
I
on the rim has reached densities which are high enough for secondary star
formation to commence.
Soft X-ray emission from within the H
I hole
is detected by a pointed
ROSAT
PSPC observation.
The emission is extended and has the same size and orientation as
the H
I shell.
These spatial properties together with a first-order spectral analysis suggest
that the emission is generated by an X-ray emitting plasma located within
the H
I shell.
However, a contribution from X-ray binaries cannot be completely ruled out
at this point.
1. Introductory Remarks
Since the discovery of large, expanding, coherent structures in the neutral
gas phase (H I) of our Galaxy (Heiles 1979) similar
structures have been observed in many nearby galaxies using powerful synthesis
radio telescopes.
These structures are most commonly referred to as H I-holes
or shells (see Brinks & Walter,
this volume and references therein, for a review
of these structures, especially in dwarf galaxies).
Several years before Heiles' paper was published, Cox and Smith (1974)
predicted that star forming regions and consecutive supernova explosions not
only put a great deal of energy into the ambient interstellar medium, in
the form of hot (T ∼ 106 - 107 K) gas,
but that this gas is long-lived.
Following their description, this should lead to an interconnecting network
of holes in the neutral interstellar medium filled with hot, coronal gas
(see also Chevalier 1974 and Weaver et al. 1977, for early analytical models
related to these structures).
In recent years, due to the advent of X-ray telescopes with sufficient
collecting area and sensitivity, this hot gas, in addition to having been
observed in our Galaxy, has now conclusively been detetected in the Magellanic
Clouds and some nearby galaxies.
For additional material, we suggest the reader to look at the contributions
in this volume by Chu,
Bomans, and Junkes, respectively, recent
theoretical work is referred to in the contributions by
Palous and
Mac Low.
Here we should like to present the detection of an expanding supergiant
H I shell within the nearby (3.2 Mpc) dwarf galaxy
IC 2574 which seems to be filled with coronal gas (as indicated by the
presence of soft X-ray emission).
A more complete presentation of the data will be presented elsewhere
(Walter et al. 1998).
2. Multi-Wavelength Observations of the Supergiant Shell
in IC 2574
2.1. Observations of the Neutral-Gas Phase
The neutral interstellar medium of IC 2574 has been studied in detail
by Walter & Brinks (1998) using H I observations
obtained with the NRAO*
Very Large Array.
An H I surface brightness map of IC 2574 is shown in
Fig. 1.
They find a total of 48 H I holes in IC 2574
most of which are expanding.
The most prominent of these holes (No. 35 in their study which is
indicated by an arrow in Fig. 1) is
the subject of this contribution.
The size of the hole is ∼1000 pc × 500 pc (which
corresponds to an effective radius reff≅350 pc) and
its radial expansion velocity is ∼25 km s-1.
The dynamical age is therefore about 1.4·107 years.
The H I mass that was present before the evacuation of
the H I hole and which must have largely accumulated on
its rim is about 2·106 Msun.
Using these numbers and correcting the H I mass for
the contribution of primordial helium, we estimate the kinetic energy of
the expanding shell to be ∼1.7·1052 erg.
Using the numerical models of Chevalier (1974), an energy input of
(2.6±1)·1053 erg by a star forming region is needed
to create this hole.
Note that the kinetic energy of the hole is of order 10% of this value.
This energy requirement corresponds to the cumulative effect of about
100 Type II SN explosions and the strong stellar winds of their
progenitors; in other words a major SF event.
*The National Radio
Astronomy Observatory (NRAO) is operated by Associated Universities, Inc.,
under cooperative agreement with the National Science Foundation.
[Click here to see Fig. 1!]
2.2. Halpha Observations
A greyscale representation of the Halpha map of IC 2574
(Walter & Brinks 1998) is given in
Fig. 2.
Note that, as indicated, the scale is different from
Fig. 1.
Although the global SF activity in IC 2574 is rather low, a ring
of H II regions is prominent in the northeast quadrant
of the galaxy (as again indicated by the arrow).
We derive that the Halpha flux on the rim is
2.3·10-12 erg cm-2 s-1
which corresponds to a luminosity of L(Halpha) =
7·105 Lsun (Lsun =
3.85·1033 erg s-1).
Note that this is almost one third of the Halpha luminosity
of 30 Dor, which is believed to be the most active SF region in
the Local Group.
The average volume densities derived on the rim of the shell are
3 cm-3 for the neutral H I (southern part)
and 2 cm-3 for the ionised hydrogen in the northern part.
This suggests that most of the neutral gas component in the northern part of
the hole has been ionized by the prominent SF regions.
[Click here to see Fig. 2!]
2.3. X-ray Observations
In order to examine the X-ray properties of IC 2574, we have analysed
a ROSAT PSPC
observation towards IC 2574 (total integration time: 7.3 ks).
A full account of the ROSAT data reduction will be presented elsewhere (Walter
et al. 1998).
A contour map of the X-ray emission is overlaid with the
Halpha-emission in
Fig. 3.
This X-ray feature has an angular extent of about
45" × 30" (FWHM) and is oriented along
the H I hole.
Since the total number of photons ascribed to the source was found to be rather
low (≅60), we decided not to attempt any sophisticated spectral modelling.
Rather, we devided the ROSAT channels into only a few energy bands.
A first-order spectral analysis (after correcting the soft ROSAT bands
for galactic extinction and determining the hardness ratios HR1 and HR2)
yields that the source emits most of the X-ray photons below E =
1 keV.
This first spectral analysis, combined with the fact that the source is
extended, suggests that a high mass, or even low mass X-ray binary is
unlikely to be the source of the detected X-ray emission.
It should be noted though that a possible contribution from X-ray binaries
to the observed X-ray emission cannot be ruled out completely at this point.
Using this simple spectral analysis, we can make a first guess for the
temperature of the hot gas that probably fills the H I hole.
Using a hot plasma model by Raymond & Smith (1977)
we derive a temperature of log (T [K]) = 6.8±0.3.
The rather large error is due to the uncertainties of our spectral analysis
(low number of photons) and the lack of a good estimate of the metallicity
in that region of IC 2574.
For a plasma of log (T [K]) = 6.8±0.3 we evaluate
an emission measure of EM =
(0.65±0.15) cm-6 pc.
Assuming an extent of the line of sight through the X-ray cavity
of 700 pc yields an electron density of ne =
(0.03±0.01) cm-3.
[Click here to see Fig. 3!]
3. Global Picture
The picture which emerges from the observations described above is as follows:
about 1.4·107 years ago (the dynamical age of
the H I hole), a major SF event took place at
the center of what today shows up as a prominent H I hole.
Since its creation some 2.6·1053 erg of energy have been
deposited by the most massive stars into the ambient ISM.
About 10% of this energy is still present in the form of the kinetic energy
of the expanding shell.
The least massive stars that go off as SN are most probably still present
in the cavity since their lifetime (∼5·107 years) is
longer than the dynamical age of the hole.
In other words, there is still energy input going on which, together with
the fact that the hole hasn't suffered blow-out (because of the relatively
thick H I disk) and hence is still pressurised, contributes
to the fact that it is bright in X-rays.
Adopting a sound speed of 100 km s-1 for the coronal gas
filling the hole, we derive a time of only
∼3·106 years before a soundwave actually reaches
the rim of the shell.
The gas has therefore had plenty of time to establish a relatively uniform
distribution within the cavity.
For a plasma temperature of log (T [K]) = 6.8±0.3
and an internal density of (0.03±0.01) cm-3 we derive an
internal pressure of P = 2neT ≅
(4±2)·105 K cm-3 (this is an order of
magnitude higher than that found by Bomans et al. (1994) in LMC 4:
P = 2·104 K cm-3).
This value should be compared to the much lower pressure of the ambient warm
ionized medium (P ≅
103-104 K cm-3).
This means that it is most likely this hot gas which is still driving the
expansion of the shell.
The shell, as it is sweeping up material, has already reached the point
at which secondary star formation can take place, as evidenced by
the H II regions located along its periphery.
References
- Bomans D.J., Dennerl K., Kürster M., 1994, A&A 283, L21
- Chevalier R.A., 1974, ApJ 188, 501
- Cox D.P., Smith B.W., 1974, ApJL 189, L105
- Heiles C., 1979, ApJ 229, 533
- Raymond J.C., Smith B.W., 1977, ApJS 35, 419
- Walter F., Kerp J., Duric N., Brinks E., Klein U., 1998, ApJL, submitted
- Walter F., Brinks E., 1998, AJ, submitted
- Weaver R., McCray R., Castor J., Shapiro P., Moore R., 1977, ApJ 218, 377
Links (back/forward) to:
| First version: | 17th | August, | 1998
|
| Last update: | 28th | September, | 1998
|
Jochen M. Braun &
Tom Richtler
(E-Mail: jbraun|richtler@astro.uni-bonn.de)