As a result of advances in instrumentation and techniques, from radio through to optical wavelengths, we have before us the prospect of producing very high resolution images of a wide range of objects across this entire spectral range. This prospect, and the new knowledge and discoveries that may be anticipated from it, lie behind an upsurge in interest in high resolution imaging from the ground. Several new high angular resolution instruments for radio, infrared, and optical wavelengths are expected to come into operation before the 1991 IAU General Assembly.

Since diffraction-limited imaging with a single aperture yields angular resolution ~ λ/D, the attainment of high angular resolution with single apertures requires the construction of correspondingly large monolithic apertures, the whole surface of which must be figured to much less than a wavelength. At the longer wavelengths, it is impossible to build a sufficiently large single aperture: for example, at λ21 cm, arcsecond resolution requires an aperture of diameter ~ 50 km. At the shorter wavelengths, the atmosphere imposes a natural limit in resolution of about one arcsecond. However, another route is possible: that of using synthetic apertures to image the sky. The problem of figuring synthetic apertures is considerably simpler, and can be implemented in a computer. Synthetic apertures are now in use in many fields, e.g. radio-interferometry, radar imaging, magnetic resonance imaging. Radio-interferometric techniques developed in radio-astronomy over the past forty years are now being applied to optical and IR astronomical imaging by a number of groups.

One aim of this meeting is to tempt us to study methods of imaging in wavebands outside our own. This might seem perverse since the imaging problems in the radio and optical/IR regions appear at first sight to differ profoundly in several ways: the rapid evolution of radio imaging over the last forty years has been largely ignored in optical imaging whilst the well developed theory and measurements of atmospheric fluctuations for the optical regime have been disregarded by radio astronomers; there have been long and exacting searches for the best sites for optical telescopes, but rarely for radio telescopes; the instrumental techniques appear to differ in many and complicated ways.

At cm wavelengths aperture synthesis radio-telescopes (arrays of linked antennas which synthesize an image of the sky with high angular resolution) are now becoming the dominant astronomical research tool. Major new facilities such as the VLA are in full operation, others such as the Australia Telescope are nearing completion and a number of telescopes designed to form images in real time have been converted to operate in the aperture synthesis mode (e.g. MOST, Bologna Cross). See Napier et al. (1983) for a review of modern synthesis telescopes. The high resolution, sensitivity and freedom from confusion have led the aperture synthesis telescopes into very diverse astronomical applications.

The technique of very long baseline interferometry (VLBI) has undergone two decades of steady growth and refinement since its inception in 1967. In the beginning, only crude measurements of visibility on single baselines were possible. Now 18-station arrays have been used to produce images with dynamic ranges exceeding 2000:1; relative motions of cosmic masers have been tracked at the microarcsecond level of accuracy; and angular size measurements have been made with baseline lengths up to 2 two earth diameters with an orbiting satellite as a receiving element.

Millimeter interferometers give high-resolution (few arc sec) information on cool (100-1000 K) matter, which cannot be easily studied by other techniques. The objects include regions of high extinction (Ay ~ 10–200 mag), planetary atmospheres, outflows from forming stars, mass loss from stellar envelopes and molecular clouds in galaxies. The advantages which mm arrays have over cm arrays are the greater number of molecular spectral lines and the larger linewidths, which allow a higher sensitivity for the same angular resolution and system performance. In the continuum mm arrays can observe solar system objects, stellar winds, compact HII regions, galaxies and quasars in thermal and synchrotron emission, and thermal radiation from dust in dense clouds. The main technical challenges are (1) sensitivity, which limits the longest usable antenna spacing, (2) small field of view, and (3) atmospheric effects, which introduce phase noise. For the next 5–10 years, the main arrays will be those of the Berkeley-Illinois-Maryland collaboration, Cal Tech, Nobeyama and IRAM. There are proposals to add more antennas to existing arrays, to build new ones, and to extend their cover to higher frequencies.

There are many outstanding astrophysical problems which are best studied at meter wavelengths. However this part of the spectrum has not been fully exploited for astronomical observations for several reasons. Firstly, the arrays at cm wavelengths provide higher resolution. Secondly, the level of man-made radio interference is very high at the longer wavelengths in industrialised countries, although it is not so high in countries such as India. Finally, the phase variations caused by the ionospheric irregularities are quite severe at meter waves, particularly for a large array (~ λ/35 rad.km-1).

During the last three years significant results have been obtained from several operational, long baseline optical interferometers. Precision stellar angular diameters (accuracy of order 2% and better) have been reported in the infrared (DiBenedetto and Rabbia, 1987) and in the visible (Davis and Tango, 1986). Astrometric precision of order 20 milliarcsec has been demonstrated over large angles (Shao et al, in press). Spectro-spatial resolution of the disk of a Be star in the hydrogen emission line (Thom et al., 1986) suggests spectacular imaging science to come with many-telescope coherent and phased optical arrays.

Speckle interferometry is a technique which utilizes the full diffraction-limited imaging potential of ground-based telescopes. Short exposure images, or specklegrams, with an exposure time less than that of the atmospheric correlation time (~5- 50 ms) preserve the high-spatial frequency information lost in long exposure imaging. In 1970, Labeyrie computed the power spectrum of a set of specklegrams and showed that they contained diffraction-limited information. Since then the field has grown with improvements in both instrumentation and the phase recovery algorithms necessary for imaging. It has been applied at both visible and near-infrared wavelengths although, until recently, the latter has used slit-scanning techniques with single pixel detectors because of the lack of array detectors. The current state of speckle interferometry has been well covered in the proceedings of two recent joint National Optical Astronomy Observatories – European Southern Observatory workshops on Interferometric Imaging in Astronomy (Oracle, 1987 & Garching, 1988).

During the last few years, two new instruments using long baseline interferometry have been constructed for high angular resolution astronomy in the mid-infrared spectral region (8-12 μxm). One called SOIRDETE-Synthese d’Overture en InfraRouge a DEux TElescopes-was built by J. Gay and his collaborators at CERGA. SOIRDETE has a fixed E-W 15 m baseline and two 1 m diameter telescopes of conventional design. This instrument obtains interference fringes by adjusting an optical-precision delay line in discrete steps to compensate for the geometrical delay of the projected baseline. The interference fringe from the source is detected using HgCdTe photodiodes. Because the instrumental delay has discrete steps a time-domain interferogram is created. This interferogram, upon Fourier transformation to the frequency domain, yields information about the spectral characteristics of the source. First fringes have recently been obtained with this instrument (Gay, 1988).

Long baseline interferometry requires the full phasing of a telescope array. Especially for future arrays with large unit telescopes active control systems are mandatory. Adaptive optics can be applied for real-time phase compensation of the individual pupils due to atmospheric distortions. Additional to phasing of the individual pupils of independently mounted telescopes, the whole array has to be phased, including pupil position corrections due to pupil foreshortening and shift effects in order to reach a reasonable phased field-of-view.

For the purposes of this talk I will take “high resolution” to mean 1 milliarcsecond (mas). Only objects of high surface brightness when averaged over 1 mas are considered; as H. McAlister will discuss stellar applications, I will speak only about nonstellar applications.

Unprecedented activity exists in very high angular resolution astronomy at optical wavelengths. Applications of telescope arrays to modern problems have been described in publications too numerous to list here. For illustrative purposes, we shall here consider an array with a maximum baseline of 350 m corresponding to a limiting resolution at the first visibility null of about 0.3 milli-arcsec at 550 nm wavelength.

The outstanding progress of radio astronomical interferometric imaging – a gain in angular resolution of 1010 in 50 years – demonstrates that once basic physical principles are established and technology developed, the field can develop with immense benefits for astronomy. Today, physics and technology give solid foundations for rapid growth of optical interferometry.

We present preliminary results of a high angular resolution-high sensitivity survey of CO(J = 2—1) line emission in M51 made with the IRAM 30 m telescope.

I wish to report on some results from mapping of molecular cloud distributions in galaxies and from tidal interaction modeling – work performed at Onsala Space Observatory and in the Astrophysics Group of Institute of Theoretical Physics, Chalmers University of Technology/University of Göteborg.

It is now well established that the environment plays an essential role in the morphology and evolution of galaxies: in particular, the HI gaseous content is often deficient for galaxies in clusters, and the deficiency increases towards the cluster center (cf for the Virgo cluster: Chamaraux et al 1980, Cayatte et al 1988, in prep). The gas is mostly deficient in the outer parts of galaxies, which considerably reduces the size of HI disks (van Gorkom & Kotanyi 1985). However the central gaseous content, usually under the form of molecular hydrogen traced by CO emission, seems normal, at least in the Virgo cluster, the only one surveyed at millimetric wavelengths (Kenney & Young 1986).

12CO(J = 1-0) observations of the central region of NGC4258 at a resolution of 17” revealed a high-density concentration of molecular gas toward the center. The molecular clump lies on a central depression in the HI bar. No clear interaction of the molecular gas with the continuum anomalous arms has been found.

Almost all galaxy properties correlate with each other to some degree; in this bounty of agreement, how are we to extract essentials? Fortunately, there is a framework in which to interpret CO correlations. Empirical studies have indicated that the fundamental galaxy properties are scale and form (Whitmore 1984), and theoretical work has implied that this follows from the physics of galaxy formation (see Lin and Pringle 1987, and references therein). It is known that the CO luminosity of galaxies is a function of both their scale (Young and Scoville 1982) and their form (Verter 1987). But the interpretation of any individual correlation is shrouded by uncertainties in the conversion from observables to source population properties.

Identifying the spiral nature of the distribution of gas in the Galaxy has been a subject of much research in the past thirty years. The position of the sun in the disk of the Galaxy presents us with a problem of perspective: how does one identify the cloud system from within the system? Longitude-velocity (l-v) diagrams have been used to try to determine the distribution of interstellar gas, but problems inherent in the methods have been pointed out previously (Burton 1971). Recent Galactic CO surveys have been used in attempts to map the distribution of molecular cloud complexes in the disk of the Galaxy (Dame, et al. 1986). Here, we use numerical simulations of the molecular cloud system in a spiral galaxy to consider the following question: to what extent can concentrations of emission in the l-v diagram (LVCs) be considered complexes of gas in the disk of the Galaxy (GMCs)?