Scientific Proposal

1. Bars in Spiral Galaxies and their importance

   The significance of spiral galaxies needs no explanation to any astronomer, and as barred spirals form around half the total spiral population, it is not necessary to explain how a study aimed at understanding the structure and dynamics of barred spirals can make a real advance in our comprehension of the Universe. In this introduction we will only highlight some of the problems implicit in our International Time Project and how they fit into the big picture.

   The dynamical problem of bar formation is intimately linked to that of disc formation, and has implications for the presence and structure of dark galactic halos. It is well known that the thin rotating discs of spirals would be unstable without the presence of massive galactic halos, or of some equivalent dynamic component. Rotating discs of gas and stars, even so, are subject to gravitational instabilities, of which the warp instability and the bar instability are among the most widespread. Modern N-body simulations readily yield stable bars. However, there is a collection of dynamical phenomena associated with bars which theory addresses semi-quantitatively, and for which excellent data sets are urgently required. These can be listed as :

   1. Bars are stated to be responsible for fueling the nuclei of at least some types of active galaxies, notably Seyferts. The mechanism is inflow of neutral gas to the centre, follewed by either strong star formation, or the supplying of a special "central engine" (or both). Direct velocity field measurements are few, and are required.
      
   2. An associated problem is the presence of resonance phenomena around the nucleus, on linear scales smaller than that of the bar, which include miniature bars, miniature spirals, and circumnuclear rings. the morphology and kinematics of these circumnuclear zones must be better observed, to pin down the theory which has been rather qualitative up to now.
      
   3. Linear theory predicts that bars should terminate at co-rotation: the radius where the rotational velocity and the pattern speed of the density wave in the disc are equal. Evidence is accumulating that this is not the case in real galaxies. We need to apply a set of tests to decide the issue.
      
   4. The interaction of a bar with the resonances in the disc is not well explored. There are new techniques for determining the zone of co-rotation, based on very careful measurements of arm morphology, or of HII region distribution. There are also new possibilities for studying ring structures associated with the Lindblad and 4:1 resonances (etc.). To these we can add global velocity distributions obtained optically. Comparison of the strengths and degrees of symmetry of density wave systems for barred and unbarred galaxies is now becoming possible, and will be tackled in this project.
      
   5. There are observational indications that a bar btimulates dynamical mixing of gas so that metallicity gradients in barred galaxy discs are reduced, or disappear. We will attempt to put this on a firmer quantitative basis for comparison with models.
   
   

   Although these problems have not been suddenly discovered, there are two powerful reasons why they should be studied now. One is that theoretical modelling has reached a stage of considerable realism, where all the dynamical factors: stars, neutral atomic and molecular gas, ionized gas, dark halo, can be well included, and in three-dimensional schemes, so that clear-cut testable predictions are now available (this would not have been the case only some three or four years ago). Secondly the balanced combination of CCD mapping photometry, in broad and narrow bands, two-dimensional emission-line spectroscopy, and classical slit spectroscopy available on the La Palma telescopes means that a self-contained project can now be expected to yield significant results. If we can add infrared measurements from Izana, and neutral gas observations from other observatories (as will be possible from the make-up of our proposing team) the attack on the problems posed will be even more powerful. In fact the team contains some of the leading theorists in the field as well as observers with a variety of skills and experience.

   In the following sections we present a more detailed accounts of the justification for the proposed observations, followed by the selection of individual objects to observed and an outline of how the observations, their analysis and their interpretation will be planned and carried out. We also give a list of participants with their institutions, and their contributing skills and any associated colleagues, as required by the terms of reference of the call for proposals for International Time Projects.
   
   

2. The Effects of Bars on Nuclear Activity

   During the past five years evidence has slowly accumulated linking activity in the nuclear zones of galaxies with the presence of a bar, or an oval distortion, in the distribution of the gravitational potential. An early study by Simkin et al. (1980) suggested the association of the presence of bars in galaxies with activity in the form of Seyfert nuclei. This study was not considered statistically convincing, i.e. not at all complete, but the underlying idea was taken seriously. Although the detailed mecanisms of nuclear activity remain in a state of partial controversy (above all the "black hole v. starburst" issue is not at all well sorted out, nor is this proposal the correct place to do so) it is physically evident that to feed any kind of outburst of energy, matter has to flow into the centre, and that for this to happen non-conservative non-central forces must operate. The existence of a non-axisymmetric structure around the nucleus foments exactly this possibilty; a stellar bar can absorb angular momentum from gas, or transfer it outwards in the more conservative case, thus allowing the gas to fall down the potential gradient into the nuclear potential well (Roberts et al. 1979, Schwarz 1985).

   There has been a sequence of studies, both observational, and theoretical, around this most interesting topic. Among the most compelling were observations of molecular gas forming "gas bars" around the nuclei of starburst galaxies, often with a circumnuclear ring of molecular gas at the centre ( Ishizuki et al. 1990, Turner & Hurt 1992). Evidence for inflow, with velocities of tens of km/s, has been cited, but the interpretation of this has been questioned (Beckman 1993) and a more plausible phenomenology is that of gas in "counterflow" with elongated orbits along the bar, leading to a net inflow at rates of order of km/s.

   The most completely studied of this type found in the literature is the Sc spiral NGC4321 (M100) which has a circumnuclear starburst. We originally studied this object for the star forming properties of its disc, and for this purpose high quality imaging in the optical (H(alpha), U, B, V, R and I bands) was obtained with excellent resolution (-0.8 arcsec seeing over the whole object), at the WHT on La Palma. To understand the parameters governing global star formation in the disc, we went on to obtain a 21 cm map with the VLA, and a map in the 1-0 line of CO, which gave information on the atomic and molecular components of the galaxy: their column of density and velocity distributions. During our analysis of the 21 cm data we were impressed with the evidence of gas in highly non-circular orbits, aligned at a certain position angle (Knapen et al. 1993a). Careful examination of our own, and previous optical images revealed the presence of a bar on a scale of 6 kpc in length, and infrared imaging, in the H band, strongly confirmed this. Intensive morphological study of the inner 1 kpc including a K-band NIR image then revealed a miniature bar, aligned with the principal bar, but separated from it by an axisymmetric zone of stars.

   The complex morphology of the circumnuclear zone, with a clear inner bar in the NIR which is not even hinted at on HST images in the optical bands, has been dealt with observationally and theoretically in two recent papers (Knapen et al. 1995a,b) by the research group of the author of the present proposal.

   We have dealt in detail with this object here because from the point of view of the International Time Project it represents a paradigm of the types of observations we will need to make in order to examine the effects of bars, and non-axisymmetric in general, in producing nuclear activity. In any sample of objects we would need to obtain:

   1. Images of the inner 10 kpc of each object in U, B, V, R and I, to pick out and photometrically map obvious, strong bars, and the limits of weaker bars and oval distortions. These could be obtained, for most objects, at the prime focus of the INT, or even (for the nearest objects whose angular resolution is a little less critical) at the JKT. Some of this work could be carried out also on the IAC-80.

   2. Images of the same zone in H(alpha) and H(beta), to pick out star forming zones, over major fractions of the discs, notably at the cusps at the ends of bars and to indicate the degree of star forming activity around the nucleus. The best instrument here would be the TAURUS camera on the WHT in imaging mode, since the instrument passes parallel light through the relevant interference filters, enabling reliable quantitative calibration (absolute H(alpha) fluxes, true dust extinction from H(alpha)/H(beta)) to be extracted, while prime focus H(alpha) narrow-band imaging at the INT is only semi-quantitative due to the converging beam. These images would also be usable for a number of projects involving massive star formation analysis over galactic disc, including (in some cases) the second part of the present proposal.

   3. A selection of the morphologically most interesting nuclear regions imaged at the best possible angular resolution, in the standard optical bands using the WHT auxiliary port, and the NOT.

   4. Velocity fields covering the bar, the circum-bar region and the circumnuclear zone, in gas via H(alpha) TAURUS velocity maps obtained at the WHT.

   5. Companions between the velocity fields of gas and stars, especially close to the nucleus. Some theoretical models predict a phase shift between the gas response and the bar potential, due mainly to the existence of periodic orbits perpendicular to the bar, between the two inner Lindblad resonances. These observations will therefore test these predictions. The proposed phase shift would enhance the gravitational torques on the gas, and will drive more quickly the gas towards the nucleus (Combes 1993). Velocity fields are obtainable via long slit spectroscopy along and perpendicular to the bar using either the ISIS on the WHT or the IDS on the INT (or both). The spectra should be obtained in a number of selected spectral ranges from the near UV (3500) to the near IR (9500) to disentangle dust effects. They would also serve as the basis for work on chemical abundance gradients, as detailed below.

   6. Near-infrared imaging of the central zones of the galaxies. The exact details would depend on the instrumentation available, but the imaging quality of the CST infrared telescope is satisfactory to image the zones where the large bars are found in the inner 10 kpc of a galaxy in J, H and K bands. Such images offer good detection possibilities for the stellar components of weak bars, and oval distortions as well as the mapping of stronger bars under low dust extinction conditions. If a suitable IR camera were available on one of the major telescopes on La Palma (WHT, INT or NOT) we would also wish to image in the near infrared (J, H, K bands) the central zones at good (<1 arcsec) angular resolution to search for bar-like or other (e.g. ring-like) structures in the stellar components there.

   7. Finally, circumnuclear near-IR spectroscopy using an instrument giving resolution of several thousand or more would offer the best way to penetrate the dust near the nucleus thus obtaining the best possible velocity fields in certain objects as well as the opportunity to study central gas concertrations, notably warm H2.

   It would be of great interest in the context of this part of the project to make complementary measurements of neutral i.e. atomic and molecular gas in the circumnuclear zones of a sub-set of our galaxies using radioastronomical techniques. In the case of HI membership in the project includes several experts in HI observations (in particular the Groningen group which implies access to the Westerbork Radio Synthesis Telescope). We would also plan to apply for VLA time for HI observations. In the case of H2, which is mapped via CO, we have experience within the group of the use of the Nobeyama milimeter interferometer and the IRAM interferometer, and would wish to make selected high angular resolution studies of circumnuclear CO, especially with the latter instrument.

   
   
   
3. The Effects of Bars on the development of Discs

   We can group these effects into two broad themes, both in terms of their astrophysical nature and the observational techniques employed.

   The first is the distribution of star formation in the galactic discs of barred galaxies. From our previous work in this field (starting with Cepa & Beckman 1990, and Knapen et al. 1992) we have detected a tendency, in grand design spirals, for the star formation regions to the distributed in two-dimensionally symmetric patterns, reflecting disc-wide resonance structures governed by density wave systems. The presence of long-lived resonances indicate that there must exist a long-lived wave-mode in these galaxise. Bars can be considered as steady density waves, or a long-lived mode in a galaxy. Corresponding resonance features are often observable in the form of nuclear, inner and outer rings at the ILR, UHR and OLR respectively. These patterns form one of the more important pieces of evidence for the existence of such systems, and also set interesting constraints on the mechanism of massive star formation. One aspect of the bisymmetric patterns is that they appear relatively strongly in galaxies with apparent absence of bars, and the symmery appears to be destroyed in barred galaxies (note that this does not refer to the arm symmetry itself, but to the pattern of star formation along the arms). It will be important to quantify this effect and set it on a statistical basis.

   The second theme is the apparent absence of metallicity gradients in barred galaxies as pointed out in a recent paper by Vila Costas & Edmunds (1992). Here again the implications are of considerable importance, because models of galaxy evolutionpredict such gradients, and they have been well observed in non-barred galaxies. In this context, Martin, Roy & Belley (1993) claimed to find some barred galaxies with metallicity gradients. Thus the initial results referred to above must be generalized if we are to deepen our understanding of gas-stellar interactions on a galactic scale. The group proposing has ample and deep experience in the chemodynamical evolution of galaxies, and the ability to make systematic use of the parameter of non-axisymmetry will be eagerly seized upon. The specific questions to be addressed here can be summarized as follows:

   1. Over how much of the disc in the abundance gradient flattened in barred spirals? This implies finding faint HII regions in the bar and as far out in the disc as possible, and taking the relevant spectra.
      
   2. How do the nuclear abundances of barred spirals compare with those of unbarred spirals? Here a difficulty is obtaining the correct radiation field near the nucleus, particularly in the case of active nuclei, but progress is being made by modifying HII region spectra with harder radiation.
      
   3. What are the gaseous motions in barred, and unbarred spirals - do the non-circular motion differ only around the bar, or in the rest of the disc? Are these uni-directional flows or do they act as a global mixing mechanism?
      
   4. How different are the star formation histories of barred and unbarred galaxies? There is a suspicious that spiral structure may in fact tend to enhance abundance gradients. How do bars counter act this? (This problem must be directly linked with the two-dimensional symmetry in the morphological theme discussed as the first section in this "disc" chapter).
      
   5. Is there any direct evidence that barred spirals (as a class or individually) have been recently involved in iteraction or merger events?
   
   

   To tackle this programme we will require the following type of observations:

   1. H(alpha) imaging of whole galaxy discs. These may often require only one on-line and a corresponding off-line exposure (for continuum subtraction), but for some systems we would need one standard and one deep frame to look fainter HII regions, while for very nearby objects we may need a multi-frame mosaic. Some of our objects in this context would be from the "nuclear activity" portion of the survey, described previously, and would not require repetition. Others would require specific observations for this disc survey.

      The observations would be divided into two classes (1) Those for which accurate photometry is essential which would require WHT TAURUS camera time and (2) Those for which global galaxy-wide information at reasonable resolution and S:N will be sufficient, which could be done using the INT prime focus, the JKT, or the IAC-80.

   2. Observations of a larger number of HII regions per galaxy, though at lower spectral resolution, taking advantage of the LDSS-II instrument on the WHT, a multi-slit instrument which enables, say, 20 regions to be observed at a time. This would be ideally efficient for large-scale chemical analysis of the disc, although it would not yield velocity information with useful precision. Use of this technique would imply previous preparation of the mask for each galaxy, based on the H(alpha) mapping described in section (a), here above. By suitable slit choice and observations of galaxies close to the meridian, any atmospheric dispersion effects on the relative emission line strengths can be kept within acceptable limits.

   3. General photometric imaging of the galaxies, to typify the stellar populations as a function of position in bulge, bar and disc, and the reveal zones of dust obscuration. This could be carried out rather slowly but entirely adequately on the JKT or IAC-80 in the standard bands (U, B, V, R, I), supplemented by near infrared imaging, as possible, with whatever instrumentation any be available, especially for galaxies with weak bars, or with heavy dust patches.

      All of this information will be supplemented by as much column density, but above all velocity mapping in neutral gas as we find appropriate. The Westerbork interferometer in HI is a reliably available instrument for HI velocity fields on galactic scales, but we may wish to use the VLA for a few selected objects to higher angular resolution. We also plan to obtain follow-up observations in CO of a few selected objects, to make estimates of the gas masses both in and outside the bar region.

      
      
      
4. Properties of Bars, and associated resonant structures

   Using direct imaging one can make a very fair approximation to the form of the gravitational potential of a barred galaxy using simple assumptions about mass-light ratios the wider the range of broad-band photometric imaging the better the approximation. From this one can carry out a set of theoretical studies: the existence and location of resonances can be determined and a predicted stationary gas-flow pattern established. These can be compared directly with the positions and shapes of observed dust-lanes which should follow the off set shocks associated with a bar (see Athanassoula 1991) as well as with observed two-dimensional radial velocity patterns in ionized and neutral gas. Developing further, a self-consistent model for the bar can then be constructed using a library of orbits, via a Schwarzschild type method; this model contains predictions of the stellar velocity field which can be checked against observations (long-slit spectra).

   Among the specific properties of bars which would be investigated are:

   1. Application of the Tremaine-Weinberg method for determining bar pattern speeds (the most effective application to date has been by Merrifield & Kuijken 1995, for NGC 936).
      
   2. How the bar's shape (tending to the rectangular when seen face-on, tending to the boxy or peanut-shaped when seen edge-on) can be explained in terms of stellar orbit theory.
      
   3. Whether there are recognizable kinematic and morphological signatures of bars in edge-on galaxies (Kuijken & Merrifield, 1994).
      
   4. Wherher weak-bars produce inner Lindblad resonances, and the dependence of the resonant radius on bar strength (shlosman & Heller 1993, Athanassoula 1991).
      
   5. Morphological correlations between bars and rings (which in principle pick out the Lindblad resonances, although this itself is not well-established, and needs further investigation) and between bars and "lenses", which are features of 50 galaxies (Kormendy 1982).
      

   We have a number of theorists in the proposing group (Athanassoula, Combes, Heller, Kuijken, Shlosman) who are eager to try out their increasingly powerful computer codes on a set of well-founded homogeneously obtained data. The optical observations for this selection: multi-band broad-band photometric imaging, long-slit spectroscopy, and two-dimensional emission spectroscopy, are all implicit in the observations explained in the previous sections, and do not imply extra types of observing, rather the systematic use of the observations obtained. The supplementary HI and CO observations will be also employed for the dynamical studies presented in this section.

   The direct support received for the project from members of leading European astronomical institutions (and the general interest shown also by colleagues in the United States) leads us to believe that the work proposed is considered of key importance and reinforces the view of the initiating group of investigators that it would mark a well-recognized step forward in extragalactic research. It is our determination, if we are granted the opportunity to make this a dynamic international project which would following the inspiring example of the late Mike Penston, give rise to new collaborations and new and permanent working friendships.

 

HOME