The P-band is the most recently completed observing band at the VLA. The aperture efficiency of the antennas is 0.4 and the system temperature is ∼ 130 K. The primary beam is slightly off-axis and its half-power width is steeply frequency dependent, which induces abnormal spectral indices on off-axis (≥ 1°) sources. We have searched for Zel’dovich “pancakes” and reached a level of ΔSrms ∼ 3.5 mJy for spectral channels of width Δν ∼ 100 kHz. We find no pancake with MHI ≳5 × 1014 M⊙.

We explore the application of optimal inversion techniques to astronomical data with a goal of developing a set of procedures for the determination of the three dimensional structure of astronomical sources. Astronomical data present a particularly difficult problem in inversion because: In any observation, 3 of 6 spatial and velocity dimensions are lost in projection onto the plane of the sky and the line of sight velocity. In any inversion, we would like to solve for a number of physical parameters. Generally, these parameters are closely related in their effect on the single observable, the sky brightness.

The dimensional deficiency leaves us with an unavoidably large degree of ambiguity (non-uniqueness) in any solution, while the inter-related parameters lead to a high probability of correlated errors and hence instability in the presence of to noise.

We show how constraints of symmetry and smoothness source allow us to handle an inversion with an insufficiently sampled data base and mutually dependent solution parameters (mathematically ill-posed and ill-conditioned). The constraints represent a priori information incorporated into the solution; thus very highly constrained inversions are similar to model fitting. In any case the inversion procedure provides us with quantitative statistics on the goodness of fit which may be used to assess the degree of ambiguity in a particular model, and the expected errors and cross-correlated errors on the parameters defining the source structure.

We briefly discuss the background and motivation, and outline the procedure in general terms. We refer to papers published in the Ap. J. where different aspects of the inversion are applied to observational data bases collected at the VLA.

In this paper I outline the strategy we have adopted to produce images from the Australia Telescope. Our main production system is based on AIPS, but we have made several departures from conventional strategy. These departures are amongst several paths that we are exploring in the hope that some will lead to richer astronomical pastures. I will describe the start that we have made on these explorations and our impressions of the scenery along the way.

Observations of speckles in the scattering disk of the Vela pulsar are presented and speckle techniques for studying and circumventing scattering of radio waves by the turbulent interstellar plasma are discussed. The speckle pattern contains, in a hologrammatic fashion, complete information on the structure of the radio source as well as the distribution of the scattering material. Speckle observations of interstellar scattering of radio waves are difficult because of their characteristically short timescales (≈seconds) and narrow bandwidths (≈kHz). Here, we present first observations, taken at 13 cm wavelength with elements of the SHEVE VLBI network, of speckles in interstellar scattering.

A simple but effective modification to the conventional CLEAN algorithm is suggested. This modification ensures both stability and speed when CLEAN is applied to maps containing a mixture of point sources and extended structures. The method has been successfully applied to the recently-completed sky survey at 34.5 MHz. This survey was made using the Gauribidanur T array (GEETEE) in 1-D aperture synthesis mode. Since in this case the ‘dirty beam’ (point spread function) cannot be directly computed, a method to obtain this is discussed in detail. The results of this deconvolution procedure have been encouraging in terms of reduced computing time and improved dynamic range in our maps. This algorithm should find wider application in deconvolving maps which have both extended structures and point sources.

A statistical criterion for stopping CLEAN procedure is proposed. The criterion is called WIC and an estimator of Kullback-Leibler information quantity which is a measure of the goodness of statistical models. A numerical example is given.

73.8 MHz instrumentation for the VLA is being developed and is currently installed on 8 VLA dishes. Test observations of strong radio sources have been made. We describe techniques that we have developed to analyze these test observations and we present examples of our first maps.

The existing method of producing radio maps from interferometric data set is based on the assumption that there is a two dimensional fourier transform relationship between the sky brightness and the measured visibility. In mapping of a very wide field, this basic assumption breaks down due to various practical problems, e.g. bandwidth smearing. We present a wide-field mapping technique which is aimed at resolving some of these problems. This technique has been successfully applied to our MERLIN observations of the 5C12 region covering an area of 1800”×l800”.

The VLA was used to derive a catalogue of phase–reference sources for MERLIN. These sources can also be used for other interferometer arrays such as the VLBA. The positions of the sources in the catalogue are repeatable to about 5 milliarcsec rms.

We have performed an observational test of the “Homogeneous Array” concept utilized in the MMA design, whereby the array elements are used in total power mode to provide the shortest spacings. Using the VLA in interferometric mode and a VLBA element measuring total power, we have reconstructed a mosaic image of the Crab Nebula, which fits the data adequately and is of good visual appearance.

We have been looking into the possibility to subtract out the continuum contribution in spectral line databases from the ungridded visibilities. It turns out that our simple method offers quite interesting possibilities in a lot of practical cases. In this contribution we show how and why this method works. Details of this work can be found in van Langevelde and Cotton (1990). We have tried to compare the results of this method with the mote conventional ways of subtracting the continuum. This comparison has to be done with simulations of which we show some results.

We describe a modification to CLEAN which alleviates some problems for extended sources. This is accomplished by combining the results of a number of conventional CLEAN operations, each done at a different resolution. The algorithm is called “Multi-Resolution Clean” or “MRC”. Experiments on model sources have shown that it works well even when the source is so extended that the usual CLEAN becomes impractical. For extended sources, MRC enhances the signal-to-noise ratio, resulting in an easier definition of the area of signal. Moreover, MRC is in principle faster than a standard CLEAN because less δ-functions are needed. This work was published in Astr. Ap., 200, 312.

We outline the results of numerical experiments on low SNR interferometer data. We confirm several theoretical analyses of the effect of averaging on visibility amplitude and closure phase data. However the triple product method for estimating the closure phase, which is useful in optical interferometry, does not provide a significant improvement over existing practice in radio interferometry. We introduce a way of using the triple product to estimate visibility amplitudes from arrays dominated by a single large telescope.

We describe a new software package, MIRIAD, for making multichannel images of multiple-field aperture synthesis observations. As an example of the use of the new software we present a map of the radio galaxy Cygnus A. We used the BIMA array to map Cygnus A at 86 GHz with 2.28 × 3.06 arcsec resolution in 3 fields centered on the central radio source and the radio hot spots in each radio lobe. The 3 fields were combined and deconvolved to find the positions and flux densities of the radio source components.

We describe the techniques currently used to produce linear polarization sensitive radio images at milliarcsecond resolution, and briefly discuss what will be possible in the next few years.

Phase-referencing is a technique used in VLBI to extend the signal coherence time from a few minutes to a few hours in order to enhance significantly its sensitivity. With this technique, VLBI observations of milliJansky radio sources can be conducted for high-accuracy differential astrometry as well as imaging. We describe the technique in some details and present, as an example, a submilliarcsecond differential astrometric experiment design to identify the star responsible for the weak radio emission in the binary system Algol.

A technique has been developed by which the data from the Mk3 correlator are read directly into the AIPS package. The data are Fourier transformed from delay lags into frequency channels and kept separately for each IF channel. The data are then corrected for multiand single band delays. We have used this technique to solve for fringe solutions on a reference source and then apply the solution to correct the data on the target source. In a recently analysed experiment we managed in this way to phase reference, using the European VLBI Network (EVN) at 18cm, the quasar OQ208 and the radio quasar 1422+202. These sources are separated by more than 9° on the sky.

The Band Width Synthesis (BWS) delay and delay rate observables measured by VLBI are used in astrometry to determine positions of extragalactic radio sources with a precision of 0.001”. At this level of precision, extragalactic radio sources exhibit “extended structures that induce significant effects in the measured BWS delays and delay rates. The question whether closure BWS delays and delay rates could supplement the amplitudes and closure phases for VLBI imaging is studied. Preliminary results for the radio source NRAO140 observed in July 1983 are presented.

We have made a detailed analysis of the systematic errors in the determination, from two sets of VLBI observations, of the position of the quasar 1038+528 A relative to the quasar 1038+528 B. This analysis confirms an apparent proper motion at λ=3.6cm of 26±8 μas/yr of the core of the quasar 1038+528 A relative to the quasar 1038+528 B.

We present evidence for changes in the radio structure of the jet in 3C120 between December 1980 and December 1988 derived from MERLIN observations at 1.67 GHz. The changes are consistent with outward motion of 19±2 mas (2.4 ± 0.3 mas yr−1, i.e. 3.5 ± 0.4h−1 c, for H0 = 100h km s−1 Mpc−1) in the inner jet, about 0.15 arcsec (∼ 75h−1 pc) from the core. This confirms the suggestion by Benson etal. (1988) that the jet does not slow down quickly just outside the core region usually imaged by VLBI. However, we find no evidence for motion at the position of the knot 4 arcsec (∼ 2h−1 kpc) from the nucleus as suggested by Walker et al. (1988). Our limit is 1.26 mas yr−1, i.e. <1.8h−1 c, compared with Walker et al. ’s measurement of 2.4 ± 0.8 mas yr−1 i.e. 3.7 ± 1.2h−1 c.