8. THE NEAR AND FAR FUTURE OF RADIO SURVEYS AND TELESCOPES

In this section I shall present some survey projects currently being carried out or planned, as well as some telescopes under construction or being designed. A good overview of current and planned radio astronomy facilities and recent research progress up to mid-1996 has been given in the latest Triennial Report of IAU Commision 40 (``Radio Astronomy''), at the URL sma2.harvard.edu/IAU_Com40/c40rpt/c40report.html. The next 3-year report is to become available in late 1999 at www.iau.org/div10.html. Many such projects have also been described in the proceedings volume by [Jackson & Davis (1997)].

8.1. Continuing or Planned Large-scale Surveys

On the island of Mauritius a 151MHz survey is being performed with the ``Mauritius Radiotelescope'' (MRT; [Golap et al. (1995)]), and may be regarded as the southern continuation of the MRAO 6C survey (cf. Table 1). This T-shaped array of helical antennas provides an angular resolution of 4' × 4.6' csc(z), where z is the zenith distance. The aim is to map the sky between declinations -10° and -70° to a flux limit of 200mJy, including a map of the Galactic Plane and studies of pulsars. A catalogue of ~ 10⁵ sources can be expected after completion of the survey in summer 1998 (icarus.uom.ac.mu/mrt2.html).

After the completion of WENSS, the WSRT started in late 1997 the ``WISH'' survey at 350MHz (www.nfra.nl/nfra/projects/index.htm). The aim is to survey the region of effective overlap with ESO's ``Very Large Telescope'' (VLT; www.eso.org/projects/vlt/), which is limited by the WSRT horizon and by the elongation of the synthesised beam. In order to have a minimum hour angle coverage of 4h, the area -30° -10°, |b| > 10°, or 5900 deg², will be covered. With an expected noise limit of about 3mJy (1-), and a source density of 20 per deg², WISH should detect about 120,000 sources.

The DRAO Penticton aperture synthesis array is being used to survey the northern Galactic plane at 408 and 1420MHz in the continuum, and at 1420MHz in the HI line (www.drao.nrc.ca/facilities/telescopes/sst/). The area covered is 72° < < 140°, -3° < b < +5°, and the angular resolutions are 1' and 4'. First results of this survey can be viewed at www.ras.ucalgary.ca/pilot_project.html.

At MPIfR Bonn a 1.4GHz Galactic plane survey (4° < |b| < 20°), using the Effelsberg 100-m dish in both total intensity and polarisation, is under way ([Uyaniker et al. (1998)]). Examples of how this survey will be combined with the NVSS, and with polarisation data from [Brouw & Spoelstra (1976)], have been shown by [Fürst et al. (1998)].

The first-epoch ``Molonglo Galactic Plane Survey'' (MGPS-1; [Green et al. (1998)]) at 843MHz, was obtained with the old, 70' field-of-view MOST and covers the region 245° < < 355°, |b| < 1.5°. The second-epoch Galactic plane survey (MGPS-2) is being made with the new, wide-field (2.7°) system at 843MHz, and will cover the region 240° 365°, |b| 10°. With an angular resolution of 43" × 43" csc and a noise level of 1-2mJy/beam it is expected to yield over 80,000 sources above ~ 5mJy ([Green (1997)]). As a part of SUMSS (Section 3.7), it is well under way, and its survey images can be viewed at www.physics.usyd.edu.au/astrop/MGPS/. Catalogues of sources will be prepared at a later stage.

The Hartebeesthoek Radio Astronomy Group (HartRAO) in South Africa, after having finished the 2.3GHz southern sky survey (Section 6.3.1), is planning to use its 26-m dish for an 8.4GHz survey of the southern Galactic plane in total intensity and linear polarisation ([Jonas (1998)]) at ~ 6' resolution, and for deeper 2.3GHz maps of interesting regions in the afore-mentioned 2.3GHz survey.

8.2. Very Recent Medium-Deep Multi-Waveband Source Surveys

Between 1995 and 1997, the ``AT-ESP'' continuum survey was carried out at 1.4GHz with ATCA ([Prandoni et al. (1998)]). This survey covers ~ 27 deg<sup>2</sup> near the South Galactic Pole with a uniform sensitivity of ~ 70 µJy (1). About 3000 radio sources have been detected, one third of them being sub-mJy sources. Redshifts from the ``ESO Slice Project'' (ESP) redshift survey for 3342 galaxies down to b_J ~ 19.4 ([Vettolani et al. (1998)], boas5.bo.astro.it/~cappi/esokp.html) will allow studies of the population of low-power radio galaxies and of their 3-dimensional distribution.

The VLA has been used in C-configuration to carry out a sensitive 1.4GHz survey of 4.22 deg² of the northern sky that have been surveyed also in the Far Infra-Red with the ISO satellite, as part of the ``European Large Area ISO Survey'' (ELAIS; [Ciliegi et al. (1998)], athena.ph.ic.ac.uk/). The 5 flux limit of the survey ranges from 0.14mJy (for 0.12 deg²) to 1.15mJy (for the entire 4.22 deg²). A careful comparison of the catalogue of 867 detected radio sources with the FIRST and NVSS catalogues provided insights into the reliability and resolution-dependent surface brightness effects that affect interferometric radio surveys. Cross-identification with IR and optical objects is in progress.

The ``Phoenix Deep Survey'' ([Hopkins et al. (1998)]) has used the ATCA to map a 2° diameter region centred on (, ) = J 01^h 14^m 12.2^s, -45° 44' 08". A total of 1079 sources were detected above ~ 0.2mJy (www.physics.usyd.edu.au/~ahopkins/cats). Optical identifications were proposed for half of the sources, and redshifts were measured for 135 of these. A comparison with lower resolution 843MHz MOST maps is in progress.

8.3. Extending the Frequency Range of the Radio Window

One of the very pioneers of radio astronomy, G. Reber, has been exploiting methods to observe cosmic radio emission at ~ 2MHz from the ground, even very recently. He quite successfully did so from two places in the world where the ionosphere appears to be exceptionally transparent (see [Reber (1994)] and [Reber (1995)]).

The lowest frequency observations regularly being made from the ground are done with the ``Bruny Island Radio Spectrometer'' ([Erickson (1997)]) on Bruny Island, south of Hobart (Tasmania). It is used for the study of solar bursts in the rarely observed frequency range from 3 to 20MHz. Successful observations are made down to the minimum frequency that can propagate through the ionosphere. This frequency depends upon the zenith distance of the Sun and is usually between 4 and 8MHz.

However, for many years radio astronomers have dreamt of extending the observing window to frequencies significantly below a few tens of MHz (where observations can be made more easily from the ground) to a few tens of kHz (just above the local plasma frequency of the interplanetary medium). Ionospheric absorption and refraction requires this to be done from space. The first radio astronomy at kHz frequencies, and the first radio astronomy from Space, was the ``Radio Astronomy Explorer'' (RAE; [Kaiser (1987)]), in the late 1960s and early 1970s. It consisted of a V-shaped antenna 450m in extent, making it the largest man-made structure in space. It was equipped with radiometers for 25kHz to 13.1MHz. Although no discrete Galactic or extragalactic sources were detected, very crude all-sky maps were made, and solar system phenomena studied. Since then none of the various space projects proposed have been realised. Recent plans for developing low-frequency radio astronomy, both from the ground and from space, can be viewed at rsd-www.nrl.navy.mil/7214/weiler by following the links to Low Frequency Radio Astronomy (LFRA) and associated pages, but see also the ALFA project (Section 8.7). The proceedings volume by [Kassim & Weiler (1990)] is full of ideas on technical schemes for very low-frequency radio observatories, and on possible astrophysical insights from them.

Efforts to extend the radio window to very high frequencies have been much more serious and successful in the past two decades, and have led to a whole new branch of ``mm-wave astronomy''. The multi-feed technique (Section 2.1) has seen a trend moving away from just having a single receiver in the focal plane, towards having multiple receivers there, to help speed up the data collection (as e.g. in the Parkes HI multibeam survey, Section 8.4). By building big correlators, and taking the cross-products between the different beams, the complex field distribution in the focal plane of a dish may be mapped, and by transforming that one can correct for pointing, dish deformation, etc. Arrays of, say, 32 by 32 feeds are able to ``image'' the sky in real time (see e.g. the SEQUOIA system at the FCRAO 14-m dish, donald.phast.umass.edu/~fcrao/instrumentation/). This is only possible at mm wavelengths, where the equipment is small enough to fit into the focal plane. In perhaps three years such receivers should exist at ~ 100GHz (3mm).

As mentioned in Section 3.4 there is a lack of large-area surveys at frequencies above ~ 5GHz. As [Condon (1998)] has pointed out, such surveys are made difficult since the beam solid angle of a telescope scales as ^-2 and system noise generally increases with frequency, so the time needed to survey a given area of sky rises very rapidly above 5GHz. However, a 7-beam 15GHz continuum receiver being built for the GBT (Section 8.6) could cover a 1-degree wide strip along the Galactic plane in one day, with an rms noise of ~ 2mJy. Repeating it several times would provide the first sensitive and systematic survey of variable and transient Galactic sources, such as radio stars, radio-emitting -ray sources, X-ray sources, etc.

More promising for the investigation of possible new source populations at these high frequencies (Section 3.4) may be the results from the new CMB satellites. One is the ``Microwave Astronomy Probe'' (MAP; map.gsfc.nasa.gov/html/web_site.html), expected to be launched by NASA in 2000. It will operate between 22 and 90GHz with a 1.4 × 1.6m diameter primary reflector, offering angular resolutions between 18' and 54'. The other one is PLANCK (astro.estec.esa.nl/SA-general/Projects/Planck/), to be launched by ESA in 2006 (possibly on the same bus as the ``Far InfraRed and Submillimetre Telescope'', FIRST, not to be confused with the VLA FIRST radio survey). PLANCK will have a telescope of 1.5m aperture, and it will be used with radiometers for low frequencies (30-100GHz; = 3-10mm), and with bolometers for high frequencies (100-857GHz; = 0.3-3.0mm), with angular resolutions of ~ 10' at 100GHz. Both the MAP and PLANCK missions should detect a fair number of extragalactic sources at 100GHz ( = 3mm). In fact, [Tegmark & De Oliveira-Costa (1998)] expect PLANCK to detect 40,000 discrete sources at 857GHz. A highly important by-product will be the compilation of a much denser grid of calibration sources at mm wavelengths. The vast majority of the currently known mm-wave calibrators are variable anyway.

8.4. Spectral Line and Pulsar Surveys

The Australia Telescope National Facility (ATNF) has constructed and commissioned a 21-cm multi-feed system with 13 receivers at the prime focus of the Parkes 64-m telescope ([Staveley-Smith (1997)]; www.atnf.csiro.au/research/multibeam/multibeam.html). The feeds are disposed to form beams with an angular resolution of 14' and a distance of ~ 28' between neighbouring feeds. The on-line correlator measures flux density in all 13 channels and 2 polarisations simultaneously, with a spectral resolution of 16kms^-1 and a velocity range from -1200kms^-1 to +12,700kms^-1. The Parkes multi-beam facility commenced regular observing in 1997, and a report on its status is regularly updated at www.atnf.csiro.au/research/multibeam/multibeam.html. Several major HI surveys are planned, including an ``all-sky'' survey ( +20°) with a limiting sensitivity (5, 600s) of ~ 20mJy per channel. The Zone of Avoidance (ZOA, |b| < 5°) will be covered with the same velocity range and twice the sensitivity. It will be sensitive to objects with HI mass between 10<sup>6</sup> and 10<sup>10</sup> M, depending on distance. This will be the first extensive ``blind'' survey of the 21-cm extragalactic sky. When scheduled, it is possible to watch the signal of all 13 beams almost in real time at wwwpks.atnf.csiro.au/people/multi/public_html/live/multibeam_live.html. An extension of this survey to the northern hemisphere ( +20°) will be performed with the Jodrell-Bank 76-m Mark I antenna, but only 4 receivers will be used.

This Parkes multibeam system is also being used for a sensitive wide-band continuum search for pulsars at 1.4GHz, initially limited to the zone 220° < < 20° within 5° from the Galactic plane. A first observing run in August 1997 suggested that 400 new pulsars may be found in this survey, which is expected to take ~ 100 days of telescope time at Parkes, spread over two years (ATNF Newsletter 34, p. 8, 1998).

The ``Westerbork observations of neutral Hydrogen in Irregular and SPiral galaxies'' (WHISP; www.astro.rug.nl/~whisp) is a survey to obtain WSRT maps of the distribution and velocity structure of HI in 500 to 1000 galaxies, increasing the number of galaxies with well-studied HI observations by an order of magnitude. By May 1998 about 280 galaxies had been observed, and the data had been reduced for 160 of them. HI profiles, velocity maps, and optical finding charts are now available for 150 galaxies. Eventually the data cubes and (global) parameters of all galaxies will also be made available.

The Dwingeloo 25-m dish is currently pursuing the ``Dwingeloo Obscured Galaxy Survey'' (DOGS; [Henning et al. (1998)]; www.nfra.nl/nfra/projects/dogs.htm) of the area 30° 220°; |b| 5.25°. This had led to the discovery of the nearby galaxy Dwingeloo 1 in August 1994 ([Kraan-Korteweg et al. (1994)]). After a shallow survey in the velocity range 0-4000 kms^-1 with a noise level of 175mJy per channel, a second, deeper survey is being performed to a noise level of 40mJy. The latter has so far discovered 40 galaxies in an area of 790 deg² surveyed to date.

The first results of a dual-beam HI survey with the Arecibo 305-m dish have been reported in [Rosenberg & Schneider (1998)]. In a 400 deg² area of sky 450 galaxies were detected, several of them barely visible on the Palomar Sky Survey.

Since 1990, the Nagoya University has been executing a ¹³CO(1-0) survey at 110GHz of the Galactic plane, with a 4-m mm-wave telescope. Since 1996, this telescope is operating at La Silla (Chile) to complete the southern Galactic plane ([Fukui & Yonekura (1998)]). The BIMA mm-array is currently being used to survey 44 nearby spiral galaxies in the ¹²CO(1-0) line at 6"-9" resolution ([Helfer et al. (1998)]).

8.5. CMB and Sunyaev-Zeldovich Effect

The cosmic microwave background (CMB) is a blackbody radiation of 2.73K and has its maximum near ~ 150GHz (2mm). Measurements of its angular distribution on the sky are highly important to constrain cosmological models and structure formation in the early Universe, thus the mapping of anisotropies of the CMB has become one of the most important tools in cosmology. For a summary of current CMB anisotropy experiments, see [Wilkinson (1998)] and [Bennett et al. (1997)]; the latter even lists the relevant URLs (cf. also brown.nord.nw.ru). As an example, the Cambridge ``Cosmic Anisotropy Telescope'' (CAT; [Robson et al. (1993)], www.mrao.cam.ac.uk/telescopes/cat/) has started to map such anisotropies, and it is the prototype for the future, more sensitive ``Very Small Array'' (VSA; Section 8.6.2).

The ``Sunyaev-Zeldovich'' (SZ) effect is the change of brightness temperature T<sub>B</sub> of the CMB towards regions of ``hot'' (T ~ 10⁷K) thermal plasma, typically in the cores of rich, X-ray emitting clusters of galaxies. The effect is due to the scattering of microwave photons by fast electrons, and results in a diminution of T_B below ~ 200GHz, and in an excess of T_B above that frequency. See [Birkinshaw (1998)] for a comprehensive review of past observations and the potential of these for cosmology. See [Liang & Birkinshaw (1998)] for the status and future plans for observing the Sunyaev-Zeldovich effect.

8.6. Radio Telescopes: Planned, under Construction or being Upgraded

8.6.1. Low and Intermediate Frequencies

The Arecibo observatory has emerged in early 1998 from a 2-year upgrading phase (www.naic.edu/techinfo/teltech/upgrade/upgrade.htm). Thanks to a new Gregorian reflector, the telescope has a significantly increased sensitivity.

The National Centre for Radio Astrophysics (NCRA) of the Tata Institute for Fundamental Research (TIFR, India) is nearing the completion of the ``Giant Metrewave Radio Telescope'' (GMRT) at a site about 80km north of Pune, India (www.ncra.tifr.res.in). With 30 fully steerable dishes of 45m diameter, spread over distances of up to 25km, it is the world's most powerful radio telescope operating in the frequency range 50-1500MHz with angular resolutions between 50" and 1.6". In June 1998, all 30 dishes were controllable from the central electronics building. Installation of the remaining feeds and front ends is expected in summer 1998. The digital 30-antenna correlator, combining signals from all the antennas to produce the complex visibilities over 435 baselines and 256 frequency channels, is being assembled, and it will be installed at the GMRT site also in summer 1998. The entire GMRT array should be producing astronomical images before the end of 1998.

The NRAO ``Green Bank Telescope'' (GBT; www.gb.nrao.edu/GBT/GBT.html) is to replace the former 300-ft telescope which collapsed in 1988 from metal fatigue. The GBT is a 100-m diameter single dish with an unblocked aperture, to work at frequencies from 300MHz to ~ 100GHz, with almost continuous frequency coverage. It is finishing its construction phase, and is expected to be operational in 2000 ([Vanden Bout (1998)]).

The VLA has been operating for 20 years now, and a plan for an upgrade has been discussed for several years. Some, not very recent, information may be found at the URL www.nrao.edu/vla/html/Upgrade/Upgrade_home.shtml. Among other things, larger subreflectors, more antennas, an extension of the A-array, a super-compact E-array for mosaics of large fields, and continuous frequency coverage between 1 and 50GHz are considered.

An overview of current VLBI technology and outlooks for the future of VLBI have been given in the proceedings volume by [Sasao et al. (1994)].

For several years the need for and the design of a radio telescope with a collecting area of one square kilometre have been discussed. The project is known under different names: the ``Square Kilometre Array Interferometer'' (SKAI; www.nfra.nl/skai/; [Brown (1996)]); the ``Square Kilometre Array'' (SKA; www.ras.ucalgary.ca/SKA/), and the ``1-km teleskope'' (1kT; www.atnf.csiro.au/1kT). A Chinese version under the name ``Kilometer-square Area Radio Synthesis Telescope'' (KARST; www.bao.ac.cn/bao/LT) was presented by [Peng & Nan (1998)], and contemplates the usage of spherical (Arecibo-type) natural depressions, frequently found in southwest China, by the placing of ~ 30 passive spherical reflectors, of ~ 300m diameter, in each of them. A frequency coverage of 0.2-2GHz is aimed at for such an array of reflectors.

A new design for a large radio telescope, based on several almost flat primary reflectors, has been recently proposed ([Legg (1998)]). The reflectors are adjustable in shape, and are of very long focal length. The receiver is carried by a powered, helium-filled balloon. Positional errors of the balloon are corrected either by moving the receiver feed point electronically, or by adjusting the primary reflector so as to move its focal point to follow the balloon. The telescope has the wide sky coverage needed for synthesis observations and an estimated optimum diameter of 100-300m. It would operate from decimetre to cm-wavelengths, or, with smaller panels, mm-wavelengths.

8.6.2. Where the Action is: Millimetre Telescopes and Arrays

The ``Smithsonian Submillimeter Wavelength Array'' (SMA; sma2.harvard.edu/) on Mauna Kea (Hawaii) consists of eight telescopes of 6m aperture, six of these provided by the Smithsonian Astrophysical Observatory (SAO) and two by the Astronomica Sinica Institute of Astronomy and Astrophysics (ASIAA, Taiwan). Eight receivers will cover all bands from 180 to 900GHz ( = 1.7-0.33mm). To achieve an optimised coverage of the uv plane, the antennas will be placed along the sides of Reuleaux triangles, nested in such a way that they share one side, and allow both compact and wide configurations. Baselines will range from 9 to 460m, with angular resolutions as fine as 0.1". The correlator-spectrometer with 92,160 channels will provide 0.8MHz resolution for a bandwidth of 2GHz in each of two bands. One of the SMA telescopes has had ``first light'' in spring 1998, and the full SMA is expected to be ready for observations in late 1999.

Since April 1998, the ATNF is being upgraded to become the first southern hemisphere mm-wave synthesis telescope (cf. ATNF Newsletter 35, Apr 1998). The project envisages the ATNF to be equipped with receivers for 12 and 3mm ([Norris (1998)]).

The ``Millimeter Array'' (MMA; www.mma.nrao.edu/) is a project by NRAO to build an array of 40 dishes of 8-10m diameter to operate as an aperture synthesis array at frequencies between 30 and 850GHz ( = 0.35-10mm). Array configurations will range from about 80m to 10km. It will most probably be placed in the Atacama desert in northern Chile at an altitude near 5000m, a site rivalling the South Pole in its atmospheric transparency ([Vanden Bout (1998)]).

The ``Large Southern Array'' project (LSA) is coordinated by ESO, IRAM, NFRA and Onsala Space Observatory (OSO), and it anticipates the building of a large millimetre array with a collecting area of up to 10,000 m², or roughly 10 times the collecting area of today's largest millimetre array in the world, the IRAM interferometer at the Plateau de Bure with five 15-m diameter telescopes. With baselines foreseen to extend to 10 km, the angular resolution provided by the new instrument will be that of a diffraction-limited 4-m optical telescope. Current plans are to provide the collecting area equivalent to 50-100 dishes of between 11 and 16m diameter, located on a plain above 3000m altitude. Currently only site testing data are available on the WWW (puppis.ls.eso.org/lsa/lsahome.html).

A similar project in Japan, the ``Large Millimeter and Submillimeter Array'' (LMSA; www.nro.nao.ac.jp/LMSA/lmsa.html) anticipates the building of a mm array of 50 antennas of 10m diameter each, with a collecting area of 3,900 m², to operate at frequencies between 80 and 800GHz.

The MMA, LSA and LMSA projects will be so ambitious that negotiations to join the LMSA and MMA projects, and perhaps all three of them, are under way. The name ``Atacama Array'' has been coined for such a virtual instrument (see NRAO Newsletter #73, p.1, Oct. 1997). The MMA will also pose challenging problems for data archiving, and in fact will rely on a new data storage medium to enable archiving to be feasible (cf. www.mma.nrao.edu/memos/html-memos/abstracts/abs164.html).

A comparison of current and future mm arrays is given in Table 3.

The ``Large Millimeter Telescope'' (LMT) is a 50-m antenna to be built on the slopes of the highest mountain in Mexico in the Sierra Negra, ~ 200km east of Mexico City, at an elevation of 4500m. It will operate at wavelengths between 8.5 and 35GHz ( = 0.85-3.4mm) achieving angular resolutions between 5" and 20" (see lmtsun.phast.umass.edu/).

There are plans for a 10-m sub-mm telescope at the South Pole ([Stark et al. (1998)]; cfa-www.harvard.edu/~aas/tenmeter/tenmeter.html). The South Pole has been identified as the best site for sub-mm wave astronomy from the ground. The 10-m telescope will be suitable for ``large-scale'' (1 deg²) mapping of line and continuum from sub-mm sources at mJy flux levels, at spatial resolutions from 4" to 60", and it will make arcminute scale CMB measurements.

The ``Very Small Array'' (VSA; www.jb.man.ac.uk/research/cmb/tele_descvsa.html; astro-ph/9804175), currently in the design phase, consists of a number of receivers with steerable horn antennas, forming an aperture synthesis array to work at frequencies around 30GHz ( = 10mm). It will be placed at the Teide Observatory on Tenerife (Spain) around the year 2000. The VSA will provide images of structures in the CMB, on angular scales from 10' to 2°. Such structures may be primordial, or due to the SZ effect (Section 8.5) of clusters of galaxies beyond the limit of current optical sky surveys.

The ``Degree Angular Scale Interferometer'' (DASI; astro.uchicago.edu/dasi) is designed to measure anisotropies in the CMB, and consists of 13 closely packed 20-cm diameter corrugated horns, using cooled High Electron Mobility Transistor (HEMT) amplifiers running between 26 and 36GHz. It will operate at the South Pole by late 1999. A sister instrument, the ``Cosmic microwave Background Interferometer'' (CBI; astro.caltech.edu/~tjp/CBI/) will be located at high altitude in northern Chile, and it will probe the CMB on smaller angular scales.

8.7. Space Projects

Radioastron ([Kardashev (1997)]; www.asc.rssi.ru/radioastron/) is an international space VLBI project led by the ``Astro Space Center'' of the Lebedev Physical Institute in Moscow, Russia. Its key element is an orbital radio telescope that consists of a deployable 10-m reflector made of carbon fiber petals. It will have an overall rms surface accuracy of 0.5mm, and operate at frequencies of 0.33, 1.66, 4.83 and 22.2 GHz. It is planned to be launched in 2000-2002 on a Proton rocket, into a highly elliptical Earth orbit with an apogee of over 80000 km.

The ``Swedish-French-Canadian-Finnish Sub-mm Satellite'' (ODIN) will carry a 1.1-m antenna to work in some of the unexplored bands of the electromagnetic spectrum, e.g. around 118, 490 and 560 GHz. The main objective is to perform detailed studies of the physics and the chemistry of the interstellar medium by observing emission from key species. Among the objects to be studied are comets, planets, giant molecular clouds and nearby dark clouds, protostars, circumstellar envelopes, and star forming regions in nearby galaxies (see kurp-www.hut.fi/spectroscopy/space-projects.shtml).

A new space VLBI project, the ``Advanced Radio Interferometry between Space and Earth'' (ARISE) has recently been proposed by [Ulvestad & Linfield (1998)]. It consists in a 25-m antenna in an elliptical Earth orbit between altitudes of 5,000 and ~ 40,000 km, operating at frequencies from 5 to 90 GHz. The estimated launch date is the year 2008.

A space mission called ``Astronomical Low-Frequency Array'' (ALFA) has been proposed to map the entire sky between 30 kHz and 10 MHz. The project is in the development phase (sgra.jpl.nasa.gov/html_dj/ALFA.html) and no funding exists as yet.

The far side of the Moon has been envisaged for a long time as an ideal site for radio astronomy, due to the absence of man-made interference. Speculations on various kinds of radio observatories on the Moon can be found in the proceedings volumes by [Burns & Mendell (1988)], [Burns et al. (1989)], and [Kassim & Weiler (1990)]. Since then, the subject has been ``dead'' as there has been no sign of interest by the major space agencies in returning to the Moon within the foreseeable future.

8.8. Nomenclature and Databases

More and more astronomers rely on databases like NED, SIMBAD & LEDA, assuming they are complete and up-to-date. However, researchers should make life easier for the managers of these databases, not only by providing their results and data tables directly to them, but also by making correct references to astronomical objects in their publications. According to IAU recommendations for the designation of celestial objects outside the solar system (cdsweb.u-strasbg.fr/iau-spec.html), existing names should neither be changed nor truncated in their number of digits. Acronyms for newly detected sources, or for large surveys, should be selected carefully so as to avoid clashes with existing acronyms. The best way to guarantee this is to register a new acronym with the IAU at vizier.u-strasbg.fr/cgi-bin/DicForm. For example, in D.Levine's lectures for this winter school the meaning of FIRST is very different from that in the present paper (cf. Section 8.3). Together with the Task Group on Designations of Commission 5 of the IAU, the author is currently involved in a project to allow authors to check their preprints for consistency with current recommendations. This should not be seen just as a further obstacle for authors, but as an offer to detect possible non-conforming designations which are likely to lead to confusion when it comes to the ingestion of these data into public databases.