New technology has enabled modern aperture synthesis instruments to employ increasingly larger bandwidths. Large bandwidths can be exploited to enable a more complete coverage in the (u, v) plane and hence aid image reconstruction. This relies on the source having a structure that is invariant with observing frequency or a suitable image reconstruction algorithm to take frequency variations into account.

At present there are several radio interferometer arrays ranging from the low-resolution (0.1–1″) high-sensitivity arrays such as the Very Large Array (VLA) to the high-resolution (0.0005–0.1″) low-sensitivity arrays such as MERLIN (though MERLIN, in the UK, is really in between this broad category), EVN, VLBA and SHEVE. Combining high sensitvity and high resolution is prohibitively expensive, because to have the u–v sampling of small arrays would require more and larger antenna elements. Hence high- resolution arrays have poor u–v coverage, decreasing the sensitivity of the instrument. This has a serious effect on spectral-line work, where sensitivity has already been sacrificed in the pursuit of spectral resolution.

This conference is concerned with the very high resolution imaging of cosmic sources in many parts of the electromagnetic spectrum. Various techniques are now available and the equipment is often automated and highly sophisticated but the term ‘very high angular resolution’ is comparative. Many of the problems existed over forty years ago when the best resolving power was about half a degree and the two major radio ‘stars’ appeared to be point sources. Very high resolution imaging in those days was the struggle to reach one minute of arc and Hanbury Brown had set his sights on considerably better than one second of arc with the concept of the intensity interferometer. The dream was to achieve a resolving power comparable to that of optical telescopes.

A major enhancement of MERLIN completed in 1992 has led to a transformation of its capabilities. The seventh telescope added to the array is a 32-m telescope at Cambridge which provides high sensitivity on all the longest baselines where extra collecting area is especially beneficial. The 32-m telescope is a shaped primary-Cassegrain configuration which achieves an 80 percent aperture efficiency. A change from frequency to amplitude modulation in the microwave links increased the effective bandwidth from 9 to 30 MHz. This broadbanding enables the two hands of circular polarization to be transmitted simultaneously, each in a 15 MHz bandwidth. A further improvement in the system sensitivity was achieved by cryogenically cooling the low noise amplifiers in the front-end receivers.

The research on the high resolution image reconstruction has been carried out at the Yunnan Observatory, the Chinese Academy of Sciences, since 1983. At the early stage in 1983–1987, our research was concentrated on the Speckle Interferometry[1]. The first developed speckle camera for using the film as the recording medium was made up of an intensifier with a gain of 57,000 and an SB-408-B oscilloscope camera. In May, 1985, 144 speckle interferograms of the binary ζBoo and the unresolved single star 32 Boo apiece were obtained with the speckle camera attached to the 1-M Rcc telescope, and the average power spectrum of ζBoo was then acquired by means of the Optical Fourier Transformation[2]. Soon afterwords, the optical couple of the intensifier to a RCA CCD(320×512 pixels) through a lens was used to replace the film and the oscilloscope camera, thereby making an improvement on the detectability up to 7m from 4m. In December, 1986 the improved camera attached to 1-M telescope was used to observe three binary stars, i.e. Kui23, Σ73 and Σ346AB. The speckle interferograms were processed with a computer and angular seperations of these binaries were obtained approximate to the diffraction-limited resolution of the telescope[3].

The Hat Creek mm-wavelength array, operated by the Berkeley-Illinois-Maryland Association (BIMA), is undergoing an expansion to nine telescopes. The first six of these telescopes will be operational in January, 1993. The telescopes are relatively small, 6 m in diameter, and are moved on rubber-tired dollies so it will be possible to transport them on roads in the vicinity of the main array. Initially, we plan to place two of the telescopes at outrigger sites located 0.5 and 1.0 km from the main array. Use of these outrigger telescopes will yield angular resolutions of 0.45 arc-sec at 2.7 mm wavelength and 0.21 arc-sec at 1.2 mm wavelength. When all nine telescopes are operational, reasonable image quality will be obtained with a single 12-hour observation. The maximum sidelobe amplitude depends upon source declination, but is about 16%. Figures 1, 2, and 3 show the uv coverage and beamshapes which will be achieved at various declinations.

A new meter-wave survey of sky region north of declination +30° is carried out with the Miyun 232 MHz Synthesis Radio Telescope(MSRT). The instrument, observation and some new imaging techniques are briefly described in this paper. As first result, two 8° × 8° regions centred respectively at α : 00h41m, δ: 41°12′ and α: 07h00m, δ: 35°00′ were observed and reduced. On the average 4 – 5 sources per square degree are recorded with position accuracy of 5′ / S(Jy). The accuracy of flux determination is limited by background fluctuation which is about 30 mJy. The catalogue is complete for sources with flux larger than 0.25 Jy.

We discuss new observations of the starburst galaxy NGC 7552. From optical and near–infrared colour maps we find a red, dusty circumnuclear ring. High-resolution radio mapping from the ATCA reveals the same ring, and a number of bright blobs (probably SNRs). The ring is probably associated with gas and dust which have lost angular momenta due to torques in the bar potential and settled at the inner Lindblad resonance. These circumnuclear starburst rings may be relatively common (when mapped without the obscuring affects of dust) and may play a role in collimating material of a nuclear outflow.

NGC 1566 and NGC 1672 have successfully been observed in radio continuum at 6 cm (26″ HPBW) with the Australia Telescope Compact Array (ATCA) using a 375-m array. We have detected for the first time with this telescope linearly polarized radio emission from two southern hemisphere spiral galaxies, NGC 1566 and NGC 1672.

This paper analyses fourth-order moments of polarized radiation passing through magnetoactive plasma with random irregularities both in electron density and magnetic field. We consider the new propagation effect arising from chiral properties of random magnetoplasma. That is, the lenses formed by the same irregularities of magnetic field may be characterized by refractive properties of opposite sense vis-a-vis the rotation of wave polarization vector. This produces an appearance of slight circular polarization fluctuations arising from initially nonpolarised radiation. Analysis of the polarized radiation fluctuations may allow the spatial spectrum of magnetic field irregularities to be detected. The enchanced level of circularly polarized component, and the share of fluctuations owing to magnetic field irregularities, can be readily observed at low frequencies only, say in the radiation passing through the solar chromosphere or Jovian and terrestrial ionosphere (magnitosphere).

The star Herschel 36 (Her 36) which ionizes most of the Hourglass nebula has been observed during a lunar occultation as well as by speckle interferometry in the near infrared (NIR). From the lunar occultation data the strip brightness profile has been extracted which shows that in the L′ band Her 36 is extended with a FWHM of 0.45″. Subsequent speckle interferometry revealed 11 sources in a 6′ × 6′ area at a level of 0.15″ resolution. From the comparison of the brightness profile obtained by the occultation and the strip brightness profile synthesized from the speckle image, we conclude that Her 36 is surrounded by a very young infrared cluster.

Two large VLBI surveys are currently underway which utilise the snapshot technique pioneered on the VLA. With a 12–16 telescope array three ~ 20 min snapshots are sufficient to make excellent hybrid maps. Recent advances in data analysis techniques enable surveys of several hundred sources to be undertaken and reduced in under two years.

We present the results of a pupil masking experiment using the Sun as a source. The goal of the experiment was a first proof of concept validation for Fizeau interferometric beam combination using a source that fills the field of view of the telescope. Phase diversity techniques are employed to record the phasing errors in the mask required to construct the optical transfer function (OTF) and are used to deconvolve the dirty images.

We have studied interferometric imaging in the multi-speckle mode by computer simulations. From various simulated data sets diffraction-limited images were reconstructed by the speckle masking method and the iterative building block method. The reconstructed images show the dependence of the signal-to-noise ratio on photon noise.

A program of High Angular Resolution observations of stars and their circumstellar regions using the technique of lunar occultations has been initiated at the 1.2 m telescope at Gurushikhar (24°39′ N, 72°47′ E), India. A liquid nitrogen cooled InSb detector based high speed Infrared photometer with millisecond data acquisition capabilities has been developed for the near Infrared region (1–5 μm) and eight occultations have been successfully observed in the K band (2.2 μm). The sources are (IRC No. -10578, +10013, +20034, +30094, +20190, +20200, +20073 and +00198. The diffraction pattern is clearly seen in all the observations. A convolution analysis involving the system frequency response, filter bandwidth and one dimensional source structure has been carried out to fit the data. System capability has been determined to be ~ 6 milliarcseconds.

We present deep 1.6 and 2.2 μm images of the central parsec of the Galaxy at a resolution of 0.15″. Most of the flux in earlier seeing limited images comes from about 340 unresolved stellar sources with K≤14. The IRS 16 and 13 complexes are resolved into about two dozen and half a dozen sources, a number of which are probably luminous hot stars. We confirm the presence of a blue near infrared object (K≈13) at the position of the compact radio source Sgr A∗. The spatial centroid of the source number distribution is consistent with the position of Sgr A∗ but not with a position in the IRS 16 complex. The stellar surface density in the central 10″ is very well fitted by an isothermal cluster model with a well defined core radius. The derived core radius of all 340 sources is 0.15±0.05 pc. The central stellar density is a few times 107 M⊙ pc−3. Buildup of massive stars by merging of lower mass stars and collisional disruption of giant atmospheres are very probable processes in the central 0.2 pc.

The spatial distribution of dust around a sample of well-known late-type stars has been studied with the Infrared Spatial Interferometer (ISI) located at Mt. Wilson. Currently operating with a single baseline as a heterodyne interferometer at 11.15 μm, the ISI has obtained visibility curves of these stars. Radiative transfer modeling of the visibility curves has yielded estimates of the inner radii of the dust shells, the optical depth at 11 μm, and the temperature of the dust at the inner radii. For stars in which the dust is resolved, estimates of the stellar diameter and temperature can also be made. Broadly speaking two classes of stars have been found. One class has inner radii of their dust shells very close to the photospheres of the stars themselves (3–5 stellar radii) and at a higher temperature (~ 1200 K) than previously measured. This class includes VY CMa, NML Tau, IRC +10216, and o Ceti. For the latter two the visibility curves change with the luminosity phase of the star and new dust appears to form at still smaller radii during minimum luminosity. The second class of stars has dust shells with substantially larger inner radii and very little dust close to the stars, and includes α Ori, α Sco, α Her, R Leo, and χ Cyg. This indicates sporadic production of dust and no dust formation within the last several decades.

The existence of circumstellar disks around young stellar objects like T Tauri stars is now well accepted. Such disks would have solar system sizes and, at the distance of the nearest star forming cloud, an angular diameter of 0.01 to 1 arcsecond at most, requiring very high angular resolution to be detected. Due to the nature of the emission process in circumstellar disks and to chromatic properties of ground based observations, disk imaging is expected to be more efficient in the near infrared. Also, multi-aperture interferometry in this wavelength range (1 – 10 μm) is expected to bring considerable insight into the disks properties and evolution in revealing their inner physical structure.

In this paper, we present synthetic images of circumstellar accretion disks. The images have been computed from a complete disk vertical structure model.

We report the observation of T Tau by visible speckle interferometry at the CFHT. The observations were carried out at two wavelengths simultaneously. Each image was split into two, allowing improved data processing. The 4-channel photon-counting detector CP40 was used. We have discovered that 30% of the total energy at Hα is emitted by an extended elliptical component, with a FWHM of 70 marcsec. In the neighbouring continuum, T Tau is not resolved by the 3.60 m CFHT. We briefly discuss the astrophysical consequences of this new result.

We present preliminary results from an imaging survey of bright northern Mira variables and Supergiants. Diffraction-limited images of Mira exhibit the same elongation along PA 120° as observed in previous observations. Of the other Mira variables observed, R Cas is unambiguously detected as non-spherically symmetric. Our observations of α-Her are consistent with there being a ‘hot-spot’ on its surface, similar to those we have previously detected on Betelgeuse.