Public Release of NIKA SZ Data

The NIKA camera was used to image galaxy clusters using the Sunyaev-Zel'dovich (SZ) effect. A total of six clusters have been observed, providing a pilot project to prepare future NIKA2 observations that are now undergoing at the IRAM 30-m telescope. In particular, wa have combined SZ data with X-ray data to study the thermodynamical properties of these clusters.

This webpage aims at describing the content of the released data products associated with this project. Note that, while NIKA was a dual-band instrument observing simultaneously at 150 and 260 GHz, only the 150 GHz data are released for the time being.

The NIKA SZ are publicly released on 2017 Nov 20th.


If you use the NIKA released data in a publication or in a conference, please consider citing at least

See publications list


The data are freely available

You may get

Cluster z R.A. DEC. Image Comments Data Main Reference
RX J1347.5-1145 0.452 13.0,47.0,30.575 -11.0,45.0,10.08 Map, point source model,
no transfer function and
noise realization for this cluster
Note that for this cluster, beam and bandpasses are different
file R. Adam et al.,
Astron. and Astrophys. 569 (2014) A66,
MACS J1423.8+2404 0.545 14.0,23.0,47.908 +24.0,04.0,42.6 Map, noise realization,
point source model, transfer function, beam and bandpass
file R. Adam et al.,
Astron. and Astrophys. 586 (2016) A122
MACS J0717.5+3745 0.546 07.0,17.0,31.740 +37.0,45.0,30.73 Map, kSZ-model-substracted map, noise realization,
point source model, transfer function, beam and bandpass
file R. Adam et al.,
Astron. and Astrophys. 598 (2017) A115
R. Adam et al.,
Astron. and Astrophys. 606 (2017) A64
PSZ1 G046.13+30.75 0.569 17.0,17.0,05.780 +24.0,04.0,26.00 Map, noise realization,
point source model, transfer function, beam and bandpass
file -
PSZ1 G045.85+57.71 0.611 15.0,18.0,20.811 +29.0,27.0,39.10 map, noise realization,
point source model, transfer function, beam and bandpass
file F. Ruppin et al.,
Astron. and Astrophys. 597 (2017) A110
CL J1226.9+3332 0.888 12.0,26.0,58.454 +33.0,32.0,48.36 map, noise realization,
point source model, transfer function, beam and bandpass
file R. Adam et al.,
Astron. and Astrophys. 576 (2015) A12
C. Romero et al.,


Explanatory supplement

150 GHz Maps
The NIKA maps are computed from raw time ordered data using the NIKA processing pipeline. See Catalano et al (2014), Adam et al (2015) and Adam et al (2014) for the main steps of the data reduction. The maps are given in units of surface brightness (mJy/beam). The data are projected on 151x151 pixels grids, where the size of the pixels are 2x2 arcsec. Images include standard astrometric metadata in the header.

Best-fit Point Source Model
The NIKA cluster fields contain non negligible radio and sub-millimeter sources identified either from NIKA data themselves, from the literature, or from Herschel data. For each cluster, we provide a point source model extracted from jointly fitting NIKA with higher (Herschel) and lower (literature) frequency data. See Adam et al (2016) for the procedure. We stress that the flux of each source in the correction maps are not necessarily well constrained, depending on the available external dataset. In addition, it depends on the underlying model that we assume for the SED of each object (either grey body or power law for infrared and radio sources, respectively). Therefore, we recommend to use this model only to quantify the impact of contaminant galaxies, but not to take it as the true contaminant signal.

Transfer function
Because of the scanning strategy and the necessary processing to subtract the correlated noise from the time ordered data, the NIKA maps are affected by filtering. This is true in particular for scales beyond the size of the field of view (~2 arcmin). The amount of filtering is estimated as described in Adam et al (2015). In brief, we compare the power spectrum of the map of simulated data before and after processing them through the NIKA pipeline. We provide the azimuthally averaged transfer function, as a function of wave number, for each cluster. We note that NIKA data are not sensitive to the zero level surface brightness of the sky. Therefore, the transfer function at wavenumber k=0 is equal to 0. The error bars provide the dispersion of the transfer function measured on individual scans, and thus, are sensitive to variation of atmospheric conditions during the data acquisition. See figure below for an example of the transfer function. We also provide the following IDL script to be used in order to apply the transfer function to non filtered data, to account for the effects of the reduction. This is to make sure that the definition of the angular frequency, k, is the same when computing and applying the transfer function onto maps. The raw NIKA maps can also be deconvolved in a similar way but the noise (in particular on the edge of the map) will drastically increase.

The beam profile corresponds to the surface brightness profile of Uranus, normalized by the amplitude of the gaussian best fit of the corresponding map (see figure below). The angular size of Uranus is typically about 3.5 arcsec, and it is neglected as it broadens the main beam by only about 0.7%.

NIKA Transfer function NIKA Beam NIKA Bandpass

Bandpass and Unit conversion
We provide the NIKA transmission as a function of frequency, obtained in the case of diffuse emission. See Figure above The following tables provide the conversion from Jy/beam to Compton parameter for different gas temperature (see Adam et al (2017) for more details).

Te (keV) I0 f(260 GHz) I0 f(150 GHz) I0 g(260 GHz) I0 g(150 GHz)
1 3.76 -11.63 7.47 12.40
5 3.31 -11.34 7.30 12.03
10 2.83 -11.00 7.11 11.62
15 2.43 -10.71 6.94 11.24
20 2.06 -10.38 6.81 10.9
25 1.76 -10.17 6.75 10.58

We also provide color correction for typical SED, both for extended emission and point sources. The assumed spectra follow f(ν) ∝ να

α Point source,
pwv=2 mm,
150 GHz
Point source,
pwv=5 mm,
150 GHz
Diffuse source,
pwv=2 mm,
150 GHz
Diffuse source,
pwv=2 mm,
150 GHz
Point source,
pwv=2 mm,
260 GHz
Point source,
pwv=5 mm,
260 GHz
Diffuse source,
pwv=2 mm,
260 GHz
Diffuse source,
pwv=2 mm,
260 GHz
-3 0.94 0.92 0.99 0.97 0.80 0.78 0.85 0.83
-2 0.97 0.95 1.00 0.98 0.85 0.84 0.89 0.88
-1 0.99 0.98 1.00 0.99 0.90 0.89 0.92 0.91
0 1.00 0.99 1.00 0.99 0.95 0.94 0.95 0.94
1 1.00 1.00 0.99 0.98 0.98 0.98 0.96 0.96
2 1.00 1.00 0.97 0.97 1.01 1.01 0.97 0.98
3 0.98 0.99 0.94 0.95 1.03 1.04 0.98 0.98
4 0.96 0.98 0.91 0.92 1.04 1.05 0.97 0.98

For each cluster, we provide a set of 100 noise realizations obtained as described in Adam et al (2016). They include correlated noise as modeled from half difference maps (Jack-Knife), as well as inhomogeneities of the noise due to the scanning strategy. The noise realizations also include the expected contribution from the CIB modeled as described in Adam et al (2017). The latter is largely subdominant for most of the data at 150 GHz, except for the deepest observations of MACS J0717.5+3745 where it becomes significant. The noise contained in the NIKA maps is well described by a gaussian noise, but we stress that it is correlated across pixels of the maps. Therefore, averaging the values of nearby pixels (e.g. when smoothing) does not reduce the noise by sqrt(N).

Special case of RX J1347.5-1145
The NIKA data provided in this page have been obtained with the final NIKA setup, and are homogeneous, except in the case of the cluster RX J1347.5-1145. For this cluster, the detectors, the optics, the bandpasses and the beam were significantly different (see Adam et al (2014)). In addition, the scanning strategy was not optimal due to a mistake in the telescope control system. Therefore, the full characterization of the data was not done for RX J1347.5-1145 and we only provide the raw map in the case of this cluster. We highly encourage to use the data only in a qualitative way (e.g., overlapping contours).

Special case of MACS J0717.5+3745
The cluster MACS J0717.5+3745 is known to host a significant amount of kinetic SZ signal (Mroczkowski et al. 2012, Sayers et al. 2013, Adam et al (2017)). Therefore, we also provide a map corrected from the best fit kSZ model obtained in Adam et al (2017). However, we stress that this models remains poorly constrained and is affected by large degeneracies between the cluster gas line-of-sight velocity and the optical depth. Therefore, in the context of thermal SZ studies, we encourage to use the kSZ corrected map to test the stability of any results, with respect to the kSZ contamination, but not to take the kSZ corrected map as the truth.

pdf version of the explanatory supplement.

Contact : Remi Adam
Last update 5/10/2023