Developments in the technologies of microwave background observation are continuing, so that there is every reason to expect that all clusters of galaxies with luminous X-ray emitting atmospheres will eventually be detected in their Sunyaev-Zel'dovich effects. Cm-wave measurements, with traditional single-dish telescopes and radiometers, are unlikely to be as effective, in the long run, as mm-wave measurements using bolometers simply because many strong X-ray clusters also contain bright radio sources whose extended emission will not easily be avoided at cm wavelengths. Nevertheless, radiometric surveys will be increasingly good at locating Sunyaev-Zel'dovich effects as arrays of receivers become more common and the bandwidths and noise temperatures of radiometers continue to improve. Over the next few years I expect the most spectacular improvements in type of Sunyaev-Zel'dovich effect work to emerge from spectral measurements of the Sunyaev-Zel'dovich effects (with the principal aim of setting limits to the velocities of clusters of galaxies) and from interferometric mapping of clusters, and indeed of the CMBR itself, using optimized interferometers.
A possible design for such an optimized array, tuned for work on clusters at redshifts 0.1, would provide µK sensitivity, a full-resolution synthesized beam 30 arcsec, and good sensitivity to angular scales 5 arcmin. For operation at cm wavelengths, this requires antennas of 10 m diameter or less, baselines from 10-100 m, and sufficient antennas simultaneously present that high sensitivity is attained rapidly and so that radio source contamination can be well mapped. Such a system is similar to BIMA or the OVMMA operated at cm-wavelengths, as done by Carlstrom et al. (1996) and Patel et al. (1997), or to the planned VSA and CBI instruments. Alternatively, smaller antennas and baselines (and smaller fractional bandwidths) could be used at a wavelength of 3 mm with a dedicated microwave background mapping array. This would have the advantages of better rejection of signals from radio sources, and more leverage on the spectrum of the Sunyaev-Zel'dovich effects with moderate changes in operating frequency, but would need a good site if it is to operate efficiently.
Survey work, as is presently carried out from ground-based antennas, could be done more efficiently from satellite systems, but with a large cost. A good initial aim for a major survey would be to provide 10 µK or better sensitivity on a large set of clusters selected without orientation bias, and hence suitable for statistical interpretation of the Sunyaev-Zel'dovich effects for cluster properties and cosmological parameters. Many clusters are likely to be detected in such an unbiased fashion in the all-sky CMBR surveys that will be produced by the next generation of mapping satellites (MAP and Planck). Long-duration balloon projects (such as SOAR) should also be able to produce excellent surveys of clusters. Cross-correlation studies between CMBR maps of large fractions of the sky and cluster (extended) X-ray sources from the ROSAT survey should give good indications of the distribution of cluster properties.
It is likely to be space-based or balloon-based operation of bolometer arrays that will produce the best measurements of Sunyaev-Zel'dovich effect spectra of clusters and hence should measure the peculiar velocities of clusters (or at least the peculiar velocities of cluster gas, which might not be the same in all cases). Combined structural and spectral measurements of a cluster, coupled with X-ray spectral and mapping information, should allow the effects of primordial structure contamination of the velocity signal in the CMBR to be minimized, since it is unlikely that the primordial perturbations behind a cluster will be distributed with an angular structure that is a close match to the cluster's gas distribution. The use of a matched filter based on the X-ray data may not be effective in all cases, however, if the structure of cluster atmospheres is found to be complicated by density and temperature inhomogeneities (as is particularly likely at higher redshifts).
Obtaining these Sunyaev-Zel'dovich effect data at high signal/noise will not be useful without matching high-quality X-ray data. Fortunately, such X-ray data will be available shortly. We are already obtaining large samples of clusters of galaxies from ROSAT (Ebeling et al. 1996), and with AXAF we will be able to obtain detailed (arcsec-resolution) X-ray images of these clusters and spatially-resolved X-ray spectra. Sunyaev-Zel'dovich interferometric maps would then be a powerful indicator of structural inhomogeneities in the gas or anomalous heating (e.g., regions of clumping, perhaps in galaxy wakes). Sunyaev-Zel'dovich and X-ray data together should provide good distance measurements over a wide range of redshifts, leading to a substantial increase in the number of clusters in the Hubble diagram (Fig. 25), but the estimation of reliable Hubble constant and deceleration parameter demands an improvement in the level of systematic errors in that diagram, especially through improvements in the calibration of the Sunyaev-Zel'dovich effect data (i.e., much better absolute calibrations of the planets, and better transfer of these calibrations to secondary sources) and the X-ray detectors.
Other CMBR data on clusters of galaxies may also become available
soon. The detection of the kinematic Sunyaev-Zel'dovich effect and the
effects from the transverse motions of clusters of galaxies would
provide a full three-dimensional velocity field of clusters, allowing
the study of the evolution of this velocity field with redshift, and
providing fundamental constraints on the physics of galaxy
clustering. Observations of Sunyaev-Zel'dovich and other effects from
galaxies (or the puzzling cluster-like structures in regions of blank
sky) are likely to provide much powerful information for cosmology and
studies of clusters over the next decade or two.
This review was partially supported by NASA grants NAGW-3825 and
contract NAS8-39073, and a research grant from
PPARC. My research on the Sunyaev-Zel'dovich effects over the years has
benefited from many collaborators, especially S.F. Gull, J.P. Hughes,
A.T. Moffet, and S.M. Molnar, and the generous
assistance of observatory staff at the
Owens Valley Radio
Observatory and the Very
Large Array. I am also
grateful to J.E. Carlstrom, M. Jones, M. Joy, J.-M. Lamarre,
A.E. Lange, and R.D.E. Saunders for providing figures and information
about their continuing observations of the Sunyaev-Zel'dovich effects,
and to P. Lilje, E. Linder, Y. Rephaeli and the referee for comments
on the text and other assistance.
This review was partially supported by NASA grants NAGW-3825 and NAG5-2415, NASA contract NAS8-39073, and a research grant from PPARC. My research on the Sunyaev-Zel'dovich effects over the years has benefited from many collaborators, especially S.F. Gull, J.P. Hughes, A.T. Moffet, and S.M. Molnar, and the generous assistance of observatory staff at the Owens Valley Radio Observatory and the Very Large Array. I am also grateful to J.E. Carlstrom, M. Jones, M. Joy, J.-M. Lamarre, A.E. Lange, and R.D.E. Saunders for providing figures and information about their continuing observations of the Sunyaev-Zel'dovich effects, and to P. Lilje, E. Linder, Y. Rephaeli and the referee for comments on the text and other assistance.