Received 02nd March 1998
Abstract.
We present preliminary results on high resolution Halpha
imaging of the
Wolf-Rayet blue compact dwarf galaxy (BCDG) Mrk 1094.
This galaxy presents a bar-shape structure and is currently undergoing
a strong star formation burst distributed in several knots.
Spatially resolved photometry of the different knots indicates that star
formation seems to be propagating from the center outwards all along the bar.
We discuss the different processes that could explain these observational
facts.
In this sense, Mrk 1094 is the first BCDG in which this kind of phenomena
is detected.
1. Introduction
Dwarf galaxies and in particular blue compact ones (BCDGs) are touchstones
towards understanding some fundamental astrophysical problems, such as
star formation (SF) and chemical and dynamical evolution of galaxies in general.
Being smaller and less massive objects compared to normal galaxies, they cannot
sustain a spiral density wave neither suffer from disk instabilities,
and their SF processes are reduced to bursts spatially (Casini & Heidman
1976; Boesgaard et al. 1982) and perhaps temporarily separated (Méndez
& Esteban 1997).
Müller & Arnett (1976) proposed that the bursts are generated
by the action of shock waves from winds and supernova (SN) explosions of
massive stars.
Gerola & Seiden (1978) and Gerola et al. (1980) generalized this model to
include the effects of stochastic local SF which they called self-propagating
stochastic SF (SPSSF).
Nevertheless it is not yet clear what firstly triggers the bursts
of SF (what Gerola & Seiden named spontaneous SF probability).
In this sense some authors have proposed that SF could be induced by tidal
interactions with H I companions (Brinks 1990;
Taylor et al. 1993, 1995 and 1996).
In order to investigate all these processes, we embarked ourselves on a
detailed optical study of the BCDG Mrk 1094.
Previous Halpha imagery of the object shows a fairly small
system of ∼5 kpc in length consisting of a string of bright knots,
laid out in an S-shape, indicating recent SF (Brinks 1990).
In the same paper, Brinks reported the discovery of an H I
Low Surface Brightness companion cloud located 51 kpc to the south.
This companion galaxy has as much mass in H I as the
parent galaxy but is almost undetected in the Halpha images.
Brinks (1990) also noted that no emission at near infrared wavelenghts was
seen in the companion, indicating the absence of older stars.
Walter et al. (1997) carried out a radio and optical (B filter)
study of the galaxy and its companion concluding that although the
H I masses of both objects are comparable,
their blue luminosities differ by an order of magnitude.
These same authors carried out a dynamical study of the pair of galaxies,
showing that no dark matter is needed to explain the rotation curves of
the parent galaxy and the companion.
In this sense they also noticed that the rotation curves are striking
different suggesting that the main galaxy contains a stellar component
whereas the companion is dominated by the gas.
However, if it is assumed that both objects are bound gravitationally,
there is need for a huge amount of dark matter surrounding the objects.
Furthermore Mrk 1094 (also known as II Zw 33) is a Wolf-Rayet
(WR) galaxy (Kunth & Joubert 1985; Vacca & Conti 1992), that is,
there are many WR stars in its interior and so the star forming knots are
young.
2. Observations
Observations were carried out on 1997 February 4 at the
2.5 m NOT
telescope of the Roque de los Muchachos Observatory at
La Palma
(Canary Islands, Spain).
We obtained high-resolution CCD images using the StanCam Camera
(1024 × 1024 pixels) with a pixel size of
24 µm2 and a spatial resolution of
0.176 arcsec pixel-1.
3. Preliminary Results
The grey-scale image of Mrk 1094
(Fig. 1 left) shows a complex
structure of different knots of SF.
The most intense knots are distributed along a central bar with two fuzzy tips
at each one of its ends.
We used the IRAF FOCAS package to detect and delimit the different knots in
the continuum subtracted Halpha image.
Thirty-seven knots were detected in total
(Fig. 1 right).
The sizes and luminosities of most intense and greatest knots are similar
(even larger) to those of giant H II regions in M33
(NGC 604) and in LMC (30 Dor).
[Click here to see Fig. 1!]
In Fig. 2 we present in grey-scale the Halpha equivalent map for the galaxy.
There are significant differences in this quantity when moving from the center
knots to the external ones (specially in the direction of the bar).
This result suggests either:
a) There exists an underlying population not distributed homogeneously along
the bar (mostly in the center).
The absorption lines of the stars of this population together with their
continuum emission can produce an underestimation of the real
W (Halpha)
associated to the star forming knots.
b) the IMF of the massive ionizing stars is different for each knot.
c) SF was not strictly coeval for all the knots.
Taking into account that:
i) the U-B colour map shows also a behaviour that indicates the youngest
stars are on the edge and
ii) the EW in absorption is negligible (Cairós et al. in preparation)
we can be confident that the influence of the hypothetic underlying population
in the estimation of the Halpha equivalent widths of the
different knots can be neglected.
In Fig. 3 we present the
distance-to-center - W (Halpha) - diagram for the thirty-seven knots detected.
If we assume the same IMF for all the star forming knots, the behaviour of
the points suggests that SF could have propagated from the center (knot #20)
outwards all along the bar.
Despite the number of regions detected
(individual star forming knots)
is not very high,
some interesting results can be derived from the luminosity and
size distributions of H II regions in Mrk 1094.
The different H II regions behave in a similar way to those in Local Group
spirals and irregulars.
The points in a luminosity-diameter-diagram fit perfectly to a straight line
of slope 3, in total agree with the results of Kennicutt (1988) for near
spirals and irregulars.
This behaviour is the expected for the simplest case of constant density
radiation-bounded H II regions where the Strömgren
diameter scales as the cube root of the luminosity.
The behaviour of the points in a size distribution histogram is also similar
to the one reported by Hodge (1983) for Local Group irregulars.
Fitting an exponential law to this distribution we found a characteristic
diameter of Do=420±30 pc (r=0.97).
This value does not fit the size-absolute magnitude relation reported by
Hodge (1983).
[Click here to see Fig. 2!]
4. Discussion
Following Gallagher et al. (1984) it is possible to derive the star formation
rate (SFR) for the galaxy at three different epochs (assuming the IMF is
invariant in time and well represented by a Salpeter Function).
Assuming a M*/LB ratio of
1 Msun/LB, sun
(the value which best fits the rotation curve of the galaxy),
Walter et al. (1997) estimate a stellar mass of
3.31·109 Msun for Mrk 1094.
This value of the stellar mass converts into a value of
0.209 Msun yr-1 for the galaxy's
lifetime averaged SFR.
Using the value of the B luminosity
(LB=3.31·109 LB, sun)
given also by Walter et al. (1997) we obtained that the averaged SFR in the
last ∼109 years is 0.557 Msun yr-1.
From our own data, we can calculate the number of Lyman continuum photons
emitted by the whole galaxy, which translates into a current SFR of
2.825 Msun yr-1.
From the previous numbers it is clear that Mrk 1094 is now undergoing a
strong star forming burst.
However the current SFR cannot account on its own for the SFR averaged during
the last 109 years.
This implies strong SF has proceeded in the galaxy over a longer period than
that related to a single burst.
Applying the population synthesis models for instantaneous bursts described
by Leitherer & Heckman (1995) and considering that the effect of the
hypothetic underlying population can be neglected,
it is possible to derive the age of each burst according to its
Halpha equivalent width
(Fig. 4).
Taking into account the difference in the estimation of the age between the
most separated knots in the bar (knot #20 and knot #37) and their linear
separation it is possible to derive that the maximum velocity of the hypothetic
perturbation that propagates SF is about 700 km s-1.
[Click here to see Fig. 3!]
[Click here to see Fig. 4!]
Another interesting question to address is the bar-shape of the distribution
of the SF knots.
This could be explained if we assume that the recent bursts of SF were
produced as a consequence of the interaction with the companion galaxy
situated 50 kpc at the south in the direction of the major axis of the bar.
Using N-body simulations for tidal interactions of two galaxies
Noguchi & Ishibashi (1986) concluded that there should exist an
overabundance of bars in interacting galaxies.
They predict that the maximum activity of the bursts produced by the
interaction takes place ∼3·108 years after the
perigalacticon which is exactly the time elapsed since the perigalacticon
for our system as derived by Walter et al. (1997).
However, there are two problems in the application of these N-body
simulations models to our system.
The first one refers to the mass of the companion, which is always equal or
higher than the parent galaxy in the models and lower in our case (Walter et
al. 1997).
The other problem is the absence of tidal tails in our system.
The models predict the existence of tails at the end of the bar that produce
finally the formation of spiral arms.
These tails are less clear when the mass of the companion is higher
(i.e. three times the mass of the parent galaxy).
In this sense, we do not see any tail, apart from the two fuzzy tips at the
extremes of the bar.
References
- Boesgaard A.M., Edwards S., Heidmann J., 1982, ApJ 252, 487
- Brinks E., 1990, in 'Dynamics and Interactions of Galaxies', Wielen R.
(ed.), Springer, p.146
- Casini C., Heidmann J., 1976, A&A 47, 371
- Gallagher J.S., Hunter D.A., Tutukov A.V., 1984, ApJ 284, 544
- Gerola H., Seiden P.E., 1978, ApJ 223, 129
- Gerola H., Seiden P.E., Schulman L.S., 1980, ApJ 242, 517
- Hodge P.W., 1983, AJ 88, 1323
- Kennicutt R.C., 1988, ApJ 334, 144
- Kunth D., Joubert M., 1985, A&A 142, 411
- Leitherer C., Heckman, T.M., 1995, ApJS 96, 9
- Méndez D.I., Esteban C., 1997, RMxAA Serie Conf, in press
- Müller W.K., Arnett W.D., 1976, ApJ 210, 670
- Noguchi M., Ishibashi S., 1986, MNRAS 219, 305
- Taylor C., Brinks E., Skillman E.D., 1993, AJ 105, 128
- Taylor C., Brinks E., Grashuis R.M., Skillman E.D., 1995, ApJS 99, 427
- Taylor C., Thomas D., Brinks E., Skillman E.D., 1996, ApJS 107, 143
- Vacca W.D., Conti P.S., 1992, ApJ 401, 543
- Walter F., Brinks E., Duric N., Klein U., 1997, AJ 113, 2031
Links (back/forward) to:
| First version: | 13th | March, | 1998
|
| Last update: | 28th | September, | 1998
|
Jochen M. Braun &
Tom Richtler
(E-Mail: jbraun|richtler@astro.uni-bonn.de)