As has been realized for a while, a cosmological observation is often accounted for by a set of quite different models. While this degeneracy can be broken by several methods, one possibility is to combine with observations of the Sunyaev & Zel'dovich (SZ) effect, the distortion of the cosmic microwave background (CMB) spectrum due to the inverse-Compton scattering in high temperature intracluster electrons (Sunyaev & Zel'dovich 1972, SZ).
RXJ 1347.5-1145 (
) is the most luminous X-ray cluster of
galaxies known so far (Schindler et al. 1997). Its exceptionally high
column density of electrons at the center implies that this is one of
the most promising clusters for the detection of the first unambiguous
SZ increment signal. We performed the search for the SZ effect towards
the cluster RXJ1347-1145, at 21 and 43 GHz with NRO (Nobeyama Radio
Observatory) 45m telescope between March 3 and 15, 1998, and at 350
GHz with SCUBA installed on JCMT (J.C.Maxwell Telescope) 15m telescope
on May 30 and 31, 1998. Although there have been a few claims of the
detection of the SZ temperature increment for some clusters (e.g.
Andreani et al. 1996; Lamarre et al. 1998), they measured the total
flux alone which could be significantly contaminated by the dust in
our Galaxy and also by submm sources in the field. Therefore it is
essential to map the profile of a cluster so that one can test it
against the predicted SZ signal profile. This is the reason why we
attempted the observation with SCUBA which has a reasonable angular
resolution, 
in FWHM, with a FOV of 
.
As a matter of fact, it turned out that a radio point source exists
near the center of our target cluster. The flux of the central point
source had been determined to be 
mJy at 1.4GHz with VLA
(Condon et al. 1988), and 
mJy at 28.5GHz in the OVRO
serendipitous survey of SZ effect (J.Carlstrom, private
communication). We also observed the central source at 100 GHz with
Nobeyama Millimeter Array (NMA) between May 19th to 21st, 1998, and
found the flux to be 
mJy. The three data at 
GHz are very well approximated by a single power-law:
where the quoted errors are 
. We subtracted the contribution
of the point source according to equation (1) from
our data at 21 and 43 GHz. Although the point source might be a
variable, the degree of variability is known to be fairly small for
radio sources with steep spectrum (spectral index 
;
e.g. Eckart, Hummel & Witzel 1989).
図 1: Radial intensity profile towards RXJ1347 at 350 GHz observed at
JCMT/SCUBA. Filled circles indicate our data with 
error-bars. Dotted curve shows the PSF of 3.2 mJy source, and dashed
(solid) curves plot the SZ profiles with (without) the possible
point source contribution, respectively, using the best-fit
parameters in the X-ray observation.
The observation of the cluster at 350 GHz with SCUBA was performed on
May 30 and 31, 1998. The raw data were processed with REMSKY in SURF
package to remove spatially correlated sky-noise. Since REMSKY
systematically changes the base-level of the map up to the sky-noise
level, the resulting zero-level is uncertain depending on the sky
condition. We estimated 1
error of our base-level or DC
offset, 
, to be as large as 
mJy/beam; the sky
condition during our observation was not so good (the zenith optical
depth at 350 GHz ranged around 
). Also we
extracted 7 spurious contamination sources above a threshold of
3
using SExtractor package. The radially averaged profile of
the cluster at 350 GHz after subtracting the sources and the
uncertain DC level is plotted in Figure 1. Also plotted
in the figure are the SZ intensity profiles at 350 GHz predicted from
the latest X-ray observation with the point source of 
and 3.2 mJy, as well as the point spread function of the 3.2 mJy
source. Since normal radio sources often show a steep decline of flux
at higher (submillimeter), we regard the value, 3.2 mJy, extrapolated
from equation (1) as the upper limit of the flux.
Again it is clear that the extended feature of the observed intensity
profile cannot be described by the point spread function alone.
To express the result in a statistical and quantitative manner, we
carried out a 
fit to the SZ profiles at 21 and 350 GHz
simultaneously on the basis of the 
-model with taking a central
y-parameter, 
, and the angular core radius, 
, as free parameters. The results are summarized in Figure
2. The agreement with the X-ray observation is
remarkable and reassuring, and we conclude that the present data
provide the strongest and most convincing case for the detection of
the submm SZ signal from the cluster. Detailed analysis taking
account of various astrophysical effects and the cosmological
implications will be presented elsewhere (Komatsu et al. 1999a,b).
図 2: Confidence contours on the y-parameter and the core radius
from combined data analysis at 21 and 350 GHz. The
cross indicates the parameters determined from X-ray observations by
ROSAT and ASCA satellites.
