next up previous
Next: Scientific Results in Press Up: IRAM Newsletter 56 (August 2003) Previous: VLBI Observations and Call


VLBI: what changes compared to a local interferometer?

Many IRAM users are familiar how the Plateau de Bure Interferometer works. There are some important differences in VLBI mode, which concerns some technical aspects but also the science which can be done. Please see the web page by T. Krichbaum (first reference of the next section ``some useful web pages'') for detailed technical information. The following points are a small introduction for observers which are not familiar with VLBI.

When operating in VLBI mode, the Plateau de Bure correlator operates in ``phased array mode'': it adds up the local baseline amplitude and phases, so that the VLBI terminal can register the signal as if it came from a single dish antenna (of an equivalent surface close to the whole interferometer).

The observations themselves are made synchronously on all participating observatories. At each telescope the so called VLBI field system, which is a VLBI control computer, runs the mutual VLBI schedule, which ensures that all telescopes observe the same source at the same time. Together with time stamps from the local H-maser clock, the complex data (amplitude and phase) are formatted, digitized and then written on magnetic tape. For each experiment only a limited amount of recording tapes are available. After the experiment, the recorded data are shipped to one of the VLBI correlator centers in the world, correlated, and made available to the P.I. of the proposal. The post-correlation analysis (calibration and fringe fitting) is done using AIPS at the P.I.'s institute.

The main advantage of VLBI is that it gives a tremendously high angular resolution (current world record: $\approx 20-30 \mu$ arcsec at 1.3mm) using the method of Earth-rotation aperture synthesis. However, only a tiny fraction of the surface of the synthesized antenna is actually filled by the contributing antennas. Therefore it is only possible to observe sources with relatively high surface brightness; brightness temperatures TB ( $\propto 10^{12}
S_{\rm Jy} \nu_{\rm GHz}^{-2} \theta_{\rm mas}^{-2}$ [K]) must be at least 108 - 109 Kelvin or 0.1-0.4 Jansky on spatial scales close to the synthesized beam size of typically 0.05 - 0.1 mas at 3 mm wavelengths (extended emission is suppressed as in a local interferometer).

Typical targets for mm VLBI are therefore quasars and masers and other compact, mainly non-thermally radiating radio sources. In the course of months the spatial resolution allows to study the evolution of maser spots around stars and details of quasar jet evolution.

Observing Frequencies: The available frequency range is (84 - 95 GHz). In standard mode, observations are done close to 86 GHz, including the SiO $(2 \rightarrow 1)$ line at 86.243 GHz in the recording bandpass. Since not all stations can tune easily to all frequencies in the aforementioned 3mm band, the proposers are ask to contact the technical VLBI advisors (see below), if they wish to observe at non-standard frequencies. In any case the frequency flexibility and the number of possible re-tunings per observing session is limited (Doppler-tracking of course is no problem).

Observing bandwidths are currently limited by the rate with which the data can be written on tape. The MK4 system offers recording rates of up to 256 Mbits/sec, corresponding to a bandwidth of up to 128 MHz. The new MK5 system offers recording with a recording rate of up to 1Gbit/s (in Europe) and up to 512 Mbit/s at the VLBA. These new recording modes are still under development.

The field of view of VLBI observations is limited by the time averaging which has to be applied, and is considerably smaller than the primary beam size of the participating antennas. In VLBI only a small field of view of some tens of synthesized beams is possible. At 3mm wavelengths this corresponds typically to a few milli-arcseconds. See the CMVA web links below for more details on the theory.

Proposal Preparation: VLBI demands observing time from many contributing institutes, which makes it more ``expensive'' than institute-specific time. Please prepare your proposals carefully with a good scientific justification and avoid last-minute submissions (The VLBA account may bounce large e-mail proposals and give FTP submission details).

Reduction Software: VLBI data is typically reduced in AIPS (Astronomical Image Processing System,

Some useful web pages

Web pages with more details on VLBI are currently under construction. The list below gives a first overview but is not exhaustive.

Michael BREMER

next up previous
Next: Scientific Results in Press Up: IRAM Newsletter 56 (August 2003) Previous: VLBI Observations and Call