Документ взят из кэша поисковой машины. Адрес
оригинального документа
: http://www.atnf.csiro.au/people/Baerbel.Koribalski/min_lecture/node27.html
Дата изменения: Unknown Дата индексирования: Mon Feb 4 07:38:44 2013 Кодировка: Поисковые слова: microwave background |
Since bars, which strongly influence the gas kinematics in the disk of a galaxy, can be triggered or enhanced by galaxy interactions (Athanassoula 1990; Gerin, Combes & Athanassoula 1990), it is necessary to study the environment of galaxies as well as their individual gas dynamics. Combes et al. (1990) show that the growing time of a bar instability (which is severely affected by various galaxy properties like e.g., the gas mass and the bulge-to-disk mass ratio) is at least 0.5109 yr. When the bar has settled down, after about 1-1.5109 yr, it rotates with a high velocity. With time, the bar length grows and its pattern speed slows down slightly. All of the nearby starburst galaxies discussed here are barred and most of them show signs of tidal interaction (e.g., disturbed velocity fields, tidal tails, HI bridges). Some of the best examples are NGC 3034 (M 82), NGC 6221, and NGC 7582. Thus, it is very likely that galaxy interaction plays an important role in the formation of bars.
Whereas in the distant ultra-luminous galaxies like Arp 220, NGC 6240, and others (see Table 4) strong tidal interactions or merging are responsible for the central activity, the less violent starbursts in nearby galaxies like NGC 253 and NGC 1808 (see Table 2) are possibly caused by bars, or oval distortions, in the disk (Combes 1988; Heckman 1990).
The best way to analyse the overall gas dynamics in galaxies is by observing
the neutral hydrogen gas. The HI velocity field, in particular, can be
used to study the streaming motions around the bar. A good example is the
barred starburst galaxy NGC 1808. Its bar is very prominent in the H
line, which is concentrated in numerous clumps along a nearly straight line
extending
3 kpc to both sides of the nucleus. Whereas it is relatively
easy to obtain velocities for the individual clumps, it is difficult to measure
the velocity field of the diffuse ionized gas surrounding the bar. Although
the bar is not as prominent in the distribution of the neutral hydrogen gas,
the deviations from circular motions caused by the bar are immediately visible
in the HI velocity field. In fact, one can nearly see the gas flowing
toward the galaxy centre where it fuels the nuclear starburst.
Theoretical studies have shown that the gas orbits within a barred potential are highly elliptical and change their shape and size near the location of resonances (see e.g., Contopoulos & Grosbøl 1989). A characteristic radius in a barred galaxy is the co-rotation radius (CR) which appears to lie near or slightly beyond the end of the bar. Numerical simulations (e.g., Combes & Gerin 1985; Combes 1988) show that the gas is streaming outwards from CR to the outer Lindblad resonance (OLR) and inwards from CR to the inner Lindblad resonance (ILR). Outer (pseudo)-rings of galaxies, usually formed by the spiral arms, indicate the accumulation of gas at the OLR. The ILR regularly lies within the central few hundred parsec of a galaxy and high-resolution observations are needed to resolve the nuclear gas kinematics. HI absorption measurements with interferometers such as the ATCA and VLA have revealed ample evidence for such rings in starburst galaxies, some of which is summarised in the previous sections. The nuclear rings or tori are found to be rotating at very large speed, often at least as high as in the outer parts of the galaxy.
One unsolved problem for numerical simulations of gas flows in galaxies is
the transport of matter into the star-forming region. Whereas the gas flow
from the disk to the nuclear ring, is well explained by the presence of a
bar, there has been no explanation on how the gas gets inside the inner
Lindblad resonance where observations clearly show the existence of large
amounts of molecular and cold atomic gas. (For a thorough discussion on
numerical simulations of gas flows in galaxies see the chapter by Jan Palous
in this volume.) To fuel, e.g., an M 82-class starburst interstellar masses
of at least 108 M
are needed (Heckman 1990). Thus, some process
must allow a substantial fraction of the ISM of a galaxy to flow right into
the nuclear region where it generates and sustains the star formation
activity and perhaps, in a more evolved stage, feeds an active nucleus (or
a black hole). Stellar winds and supernova explosions eventually lead to the
formation of chimneys and fountains, where ionized gas breaks out of the
galactic plane, dragging neutral gas and dust (and magnetic field lines) out
to large scale-heights. The outflows in NGC 253, M 82 and NGC 1808 are
well studied examples.
Rieke, Lebofsky & Walker (1988) suggest a sequence for the evolution of a
nuclear starburst in which star formation occurs first at the nucleus and
then spreads over the whole nuclear region, accompanied by gas flowing out
from the centre. In this sequence NGC 253 is the prototype of `phase 3' where
the nuclear star formation has ended and superbubbles break into the halo. In
`phase 4' the circum-nuclear star formation continues and the given prototype
is M 82. At the end of the sequence, when star formation has ceased, stand
galaxies like M 31 (Andromeda) where the emission is dominated by old stellar
populations. M 31 is also a promising candidate for a black hole in the centre
as the rotation velocity and the velocity dispersion strongly increases in the
central few pc (see Dressler & Richstone 1988; Kormendy 1988, 1990). Maybe
our Galaxy is also at the end of this sequence showing little star-formation
activity, but large velocities in the nuclear disk (see Dame et al. 1987).