2.3. Cluster gas structure
The structures of the atmospheres of clusters of galaxies are generally studied by X-ray rather than SZ techniques at present, since the signal/noise achievable in the X-ray is far superior. However, the X-ray emissivity is proportional to the square of the electron density in the gas rather than to the electron density, so both the appearance of the X-ray image and the weighting of gas emissivity in the X-ray spectrum are heavily influenced by the centre of the cluster. The SZ effects, on the other hand, are directly proportional to electron density, so they are relatively more sensitive to the outer parts of the clusters and could be better probes of these regions. At present the quality of the SZ data is such that only relatively crude information on the structures of the atmospheres of nearby clusters is available (e.g., Lancaster 2004), but the redshift-independence of the SZ effects could make them the best probes of the structures of distant clusters. For the closest clusters, multi-wavelength studies will be needed to remove the "noise" contributed by primordial structure in the MBR.
For simple structural measurements using the thermal SZ effect no absolute calibration is needed, and the X-ray data are useful only if the conversion from the projected electron pressure profile to an electron density profile is of interest. However the extraction of the intrinsic profile from the observed profile relies on a deconvolution with the telescope beam.
2.3.1. Beamshape measurements
As discussed in Sec. 1.3.1, the telescope beamshape must be well-known for accurate interpretation of the data. There are two types of beamshape measurement that are needed to extract the best science from SZ effect data. First, since clusters generally have a large angular size, they are likely to fill a substantial fraction of the primary beam of an interferometer used to observe the cluster (essentially, this is the same condition as the requirement on the interferometer sampling for the observation to have a high efficiency). This implies that the primary beams of the antennas in the interferometer need to be well known if the SZ effect structure is to be measured well on large angular scales.
On the other hand, small-scale structures are well represented in interferometric data, and for these the beamshape is determined by the sampling of the u - v plane if the phase and amplitude errors in the data are small. Thus small-scale SZ effect structures should be well captured by interferometric observations and the deconvolutions required to find their intrinsic properties should be reliable.
Structural measurements of this type are clearly subject to confusion. In Sec. 1.7 it was shown that primordial structures in the MBR add a degree of noise to an SZ image that cannot be removed without the use of multi-frequency observations to effect a spectral separation of the thermal effect (and this brings in, again, the requirement for high-quality spectral calibration; Sec. 2.1.1). This imposes a practical limit on the angular scales of SZ effect structure which can be studied at present.
At smaller angular scales, where the confusion from MBR structures is reduced, confusion from non-thermal radio sources (principally quasars and high-redshift radio galaxies) becomes important (Fig. 1). Removal of most of the effect of these radio sources is feasible using interferometers, as discussed in Sec. 1.6.1, by the correct choice of baselines or by aperture-plane fitting. However at the lowest SZ effect flux densities it is hard to remove the confusing sources cleanly.