CL J1226.9+3332: High‑z Galaxy Cluster

• CL J1226.9+3332 is a massive galaxy cluster at z≈0.89 defined by extensive millimeter, X‑ray, lensing, and optical observations.
• High‑resolution tSZ data reveal a smooth large-scale pressure profile alongside a dynamically disturbed core with merger signatures.
• Joint multi‑probe analyses provide consistent hydrostatic and lensing mass estimates with detailed thermodynamic profiles and gas fractions of 0.13–0.15.

CL J1226.9+3332, also designated RX J1226.9+3332, is a massive galaxy cluster at $z=0.89$ or $z=0.888$ in the cited analyses. It is the highest-redshift object in the NIKA2 Sunyaev-Zel'dovich Large Program and has been observed at millimeter, X-ray, lensing, and optical wavelengths. Across these studies, the cluster is described as fairly regular on large scales but with a disturbed core, and it has served as a proof of concept for multi-wavelength studies of thermodynamics, hydrostatic mass reconstruction, hydrostatic-to-lensing bias, and galaxy evolution at high redshift [(Adam et al., 2014); (Muñoz-Echeverría et al., 2022); (Muñoz-Echeverría et al., 2021); (0910.0511)].

1. Astrophysical identification and observational context

CL J1226.9+3332 sits at $z=0.89$, with one multi-probe study quoting $z=0.888$, and is explicitly identified as the highest-redshift object in the NIKA2 SZ Large Program. For the cosmology adopted in the multi-probe hydrostatic-mass analysis, $1'$ on the sky corresponds to $\simeq 466\ \mathrm{kpc}$ at $z=0.89$, and the cluster’s characteristic scale is typically reported as $R_{500}\simeq 900$–$1000\ \mathrm{kpc}$ (Muñoz-Echeverría et al., 2022).

The system has been investigated with NIKA, NIKA2, MUSTANG, Bolocam, NOEMA, Chandra, XMM-Newton, Planck, HST/CLASH, Gemini/GMOS, and HST/ACS. This breadth of coverage has made it a recurrent target for cross-calibration of tSZ, X-ray, and lensing mass estimators, as well as for optical studies of galaxy structure and the Fundamental Plane [(Adam et al., 2014); (Muñoz-Echeverría et al., 2022); (0910.0511)].

A persistent observational theme is the coexistence of large-scale regularity with central disturbance. Large-scale X-ray surface brightness is described as fairly regular or relaxed, whereas higher-resolution tSZ, X-ray temperature, radio, and lensing data reveal a south-west substructure and other signatures of a past merger [(Adam et al., 2014); (Muñoz-Echeverría et al., 2022)].

2. Millimeter-wave and tSZ observations

High-resolution tSZ work on CL J1226.9+3332 began with NIKA on the IRAM 30-m telescope. In the full NIKA analysis, the camera operated at 150 and 260 GHz with Gaussian beams of $18.2''$ and $z=0.888$0, a field of view of $z=0.888$1, and an effective integration time of $z=0.888$2 hours on source acquired with on-the-fly raster scans. Calibration was performed on Uranus, with overall uncertainty $z=0.888$3 at 150 GHz and $z=0.888$4 at 260 GHz. The 260 GHz map showed a single bright point source, PS$z=0.888$5, located $z=0.888$6 south-east of the X-ray center, with best-fit fluxes $z=0.888$7 at 260 GHz and $z=0.888$8 at 150 GHz. After subtraction of PS$z=0.888$9, the 150 GHz map revealed a smooth tSZ decrement peaking at $z=0.89$0 per beam (Adam et al., 2014).

The earlier NIKA proceeding emphasized instrumental performance and feasibility. It reported simultaneous 150/260 GHz operation with on-sky beam sizes $z=0.89$1 and $z=0.89$2, typical per-detector sensitivities of order $z=0.89$3 at 150 GHz and $z=0.89$4 at 260 GHz, cross-linked on-the-fly raster scans, and total on-source time $z=0.89$5 at 150 GHz and $z=0.89$6 at 260 GHz. That dataset yielded a peak tSZ signal-to-noise of $z=0.89$7, with final tSZ maps on $z=0.89$8 pixels; the 150 GHz map contained the cluster decrement, while the 260 GHz map was used to trace and subtract foreground contaminants (Adam et al., 2014).

Later interferometric imaging with NOEMA extended the angular-resolution frontier. The NOEMA study used ten antennas in the most compact D configuration, with $z=0.89$9 on source, dual sidebands covering 70–78 GHz and 86–94 GHz, and a synthesized beam of $z=0.888$0 at position angle $z=0.888$1. After source subtraction, the final $z=0.888$2 continuum sensitivity was $z=0.888$3. Dirty-map inspection and spectral-cube line search identified three contaminating point sources: PS9, identified with the brightest cluster galaxy at $z=0.888$4 from the tSZ peak, and PS260–N and PS260–S, two submillimetre-galaxy line emitters at $z=0.888$5 south-east. These were modeled with uv_fit in GILDAS/MAPPING and removed directly in the uv-plane before construction of a source-clean continuum dataset (Muñoz-Echeverría et al., 17 Sep 2025).

3. Pressure reconstruction and thermodynamic structure

The tSZ analyses adopt the standard Compton-$z=0.888$6 description

$z=0.888$7

with observed temperature change

$z=0.888$8

and, to first order neglecting relativistic corrections,

$z=0.888$9

For NIKA’s bands, the proceeding quoted $1'$0 and $1'$1, so the 150 GHz channel measures the decrement while the 260 GHz channel primarily diagnoses dusty or radio contamination (Adam et al., 2014).

Radial pressure reconstruction has been performed in both parametric and non-parametric forms. The NIKA analysis modeled the three-dimensional electron pressure with a generalized NFW form,

$1'$2

while the Chandra density profile was fitted with a simplified Vikhlinin model. In the baseline PPC pressure model, the best-fit normalization and scale radius were reported as $1'$3 and $1'$4. A simultaneous MCMC fit of the tSZ plus point-source model to the 150 GHz map, together with a Planck-derived integrated signal in a $1'$5 aperture,

$1'$6

yielded full three-dimensional profiles of pressure, density, temperature, entropy, and mass (Adam et al., 2014).

That reconstruction implied a core temperature $1'$7, declining to $1'$8 at $1'$9, and an entropy law $\simeq 466\ \mathrm{kpc}$0 with flattening below $\simeq 466\ \mathrm{kpc}$1 and $\simeq 466\ \mathrm{kpc}$2. In that interpretation, the core was already thermodynamically disturbed despite the cluster’s relaxed large-scale appearance (Adam et al., 2014).

NOEMA enabled a core-to-outskirts pressure reconstruction from $\simeq 466\ \mathrm{kpc}$3 to $\simeq 466\ \mathrm{kpc}$4. In the non-parametric model, the pressure bins were fitted at $\simeq 466\ \mathrm{kpc}$5, giving $\simeq 466\ \mathrm{kpc}$6, $\simeq 466\ \mathrm{kpc}$7, $\simeq 466\ \mathrm{kpc}$8, and $\simeq 466\ \mathrm{kpc}$9, respectively. A joint gNFW fit to NOEMA, IRAM 30 m/NIKA2, MUSTANG, and Bolocam yielded a continuous $z=0.89$0 with 16–84% and 2.5–97.5% credible regions covering $z=0.89$1–$z=0.89$2 (Muñoz-Echeverría et al., 17 Sep 2025).

The cluster also became a worked example for joint X-ray–SZ forward modeling in JoXSZ. In that framework, the Abel-projected pressure profile is convolved with the beam and transfer function, X-ray emissivity is translated to count rate with XSPEC, missing data are omitted from the likelihood, and calibration or background systematics are represented by nuisance parameters. Applied to Chandra plus NIKA data, the full joint fit found $z=0.89$3 and $z=0.89$4, while the posterior on the temperature-ratio parameter implied

$z=0.89$5

with a Savage–Dickey Bayes factor $z=0.89$6, characterized as weak, inconclusive evidence for a true offset (Castagna et al., 2020).

4. Mass scale, hydrostatic reconstruction, and lensing comparison

Hydrostatic analyses use the standard equilibrium relation

$z=0.89$7

with

$z=0.89$8

For CL J1226.9+3332, published values of $z=0.89$9 span a relatively narrow range when tSZ and X-ray information are combined, but the multi-probe literature repeatedly notes that slight differences in the shape of the mass profile near $R_{500}\simeq 900$0 can substantially shift the integrated mass (Muñoz-Echeverría et al., 2022).

Analysis	Reported result	Label
NIKA + Chandra + Planck, baseline PPC	$R_{500}\simeq 900$1; $R_{500}\simeq 900$2	tSZ/X-ray/Planck
JoXSZ full joint fit, HE assumed	$R_{500}\simeq 900$3; $R_{500}\simeq 900$4	Joint forward model
SZ + X, gNFW approach	$R_{500}\simeq 900$5	Multi-probe
SZ + X, NFW backward approach	$R_{500}\simeq 900$6	Multi-probe
X-ray-only HSE	$R_{500}\simeq 900$7	Multi-probe
Lensing	$R_{500}\simeq 900$8	CLASH-based
Combined variants, proof of concept	$R_{500}\simeq 900$9; $1000\ \mathrm{kpc}$0	NIKA2 LPSZ proof of concept

The corresponding gas fractions are also consistent at the $1000\ \mathrm{kpc}$1–$1000\ \mathrm{kpc}$2 level. The NIKA+Chandra+Planck study obtained

$1000\ \mathrm{kpc}$3

whereas the JoXSZ full joint fit found

$1000\ \mathrm{kpc}$4

An earlier NIKA proceeding, using a self-similar $1000\ \mathrm{kpc}$5–$1000\ \mathrm{kpc}$6 scaling relation, estimated $1000\ \mathrm{kpc}$7 with statistical $1000\ \mathrm{kpc}$8 and $1000\ \mathrm{kpc}$9 [(Adam et al., 2014); (Castagna et al., 2020); (Adam et al., 2014)].

The hydrostatic-to-lensing bias has been treated as a central systematic question. In the exhaustive multi-probe analysis, the observational estimates were $18.2''$0 for the gNFW HSE reconstruction, $18.2''$1 for the NFW backward reconstruction, and $18.2''$2 for the X-ray-only HSE reconstruction. The range $18.2''$3–$18.2''$4 was presented as the net effect of data-set choice and mass-profile modeling. The earlier proof-of-concept analysis instead summarized the comparison as

$18.2''$5

that is, consistent with zero within 20%. Both studies emphasized that model dependence dominates instrumental systematics in the final mass budget, while transfer-function choices in the NIKA2 analysis were found to have $18.2''$6 impact on $18.2''$7 and a $18.2''$8–$18.2''$9 effect on the final mass (Muñoz-Echeverría et al., 2022, Muñoz-Echeverría et al., 2021).

5. Morphology, core state, and dynamical interpretation

The 2014 NIKA map showed a cluster that was azimuthally smooth on scales $z=0.888$00 but elongated to the south-west on $z=0.888$01–$z=0.888$02 scales. That elongation was associated with three previously reported features: an XMM/Chandra hotter south-west region, a MUSTANG 90 GHz ridge $z=0.888$03 long and $z=0.888$04 south-west of the center, and an HST lensing subclump $z=0.888$05 south-west. The interpretation given there was a minor merger that had stripped or heated core gas (Adam et al., 2014).

Subsequent multi-probe work reinforced the disturbed-core picture. XMM-Newton and Chandra reveal a hot, south-west substructure $z=0.888$06 from the X-ray peak, and optical plus LOFAR data show a secondary mass clump and a giant radio halo, presented as signatures of a past merger. At the same time, the X-ray surface brightness remained fairly regular on larger scales, preserving the recurrent description of a globally regular cluster with central dynamical complexity (Muñoz-Echeverría et al., 2022).

The NOEMA core mapping sharpened this interpretation. It provided the first tSZ-based map of the core, $z=0.888$07, in a $z=0.888$08 cluster and reported that the pressure profile is flat in the core of the cluster. The best-fit gNFW inner slope was $z=0.888$09, corresponding to a shallow pressure gradient inside $z=0.888$10. Residual imaginary visibilities pointed to departures from circular symmetry, including sloshing and substructures, and the combined evidence from weak-lensing mass subclumps, X-ray temperature asymmetry, and diffuse radio emission was taken to confirm that CL J1226.9+3332 is dynamically disturbed. The same study described the center as having a “puffed-up” pressure distribution, consistent with ICM sloshing or an early-stage merger, and suggested possible interplay between intracluster medium and circumgalactic medium around the brightest cluster galaxy (Muñoz-Echeverría et al., 17 Sep 2025).

An earlier NIKA proceeding had compared the pressure profile to the low-$z=0.888$11 “universal” gNFW template of Arnaud et al. (2010) and reported that CL J1226.9+3332 appears marginally more compact in the core, with its location on the $z=0.888$12–$z=0.888$13 plane consistent, within $z=0.888$14, with the local scaling relation. A plausible implication is that increasingly resolved millimeter data shifted the emphasis from global scaling consistency toward explicit core disturbance and non-circularity (Adam et al., 2014).

6. Galaxy population, structural parameters, and the Fundamental Plane

Optical work on RX J1226.9+3332 places the cluster in a second, complementary context: galaxy structural evolution. Gemini/GMOS spectroscopy targeted $z=0.888$15–$z=0.888$16 objects in the field, and the extended sample comprised 87 objects. HST/ACS imaging used four WFC pointings in F606W and F814W, each with $z=0.888$17 exposures, covering about $z=0.888$18 per pointing at $z=0.888$19 pixel scale. Structural parameters were measured for 112 galaxies, of which 54 were spectroscopically confirmed cluster members. Targets were chosen independently of morphology to avoid progenitor bias (0910.0511).

Surface-brightness modeling employed GALFIT with both de Vaucouleurs and Sérsic profiles and a $z=0.888$20 drizzled Tiny Tim PSF. For the 54 confirmed members, the half-light radius ranged from $z=0.888$21 to $z=0.888$22, corresponding to $z=0.888$23–$z=0.888$24 for $z=0.888$25, $z=0.888$26, $z=0.888$27, with median $z=0.888$28 and $z=0.888$29 scatter $z=0.888$30. The effective surface brightness ranged from $z=0.888$31 to $z=0.888$32, with median $z=0.888$33 and scatter $z=0.888$34. Sérsic indices spanned $z=0.888$35 to $z=0.888$36, with mean $z=0.888$37, median $z=0.888$38, rms $z=0.888$39, and fraction with $z=0.888$40 of about $z=0.888$41. Simulations showed that forcing an $z=0.888$42 fit on a true Sérsic $z=0.888$43 galaxy gives $z=0.888$44 and $z=0.888$45, but the Fundamental-Plane combination $z=0.888$46 was affected only at the level $z=0.888$47 (0910.0511).

For the early-type members, the Fundamental Plane was reported as

$z=0.888$48

The plane remained tight, with rms in $z=0.888$49 of $z=0.888$50 dex, but its slope differed from the local Coma relation. The quoted interpretation was “downsizing”: lower-mass galaxies experienced more recent luminosity evolution than the most massive systems. The zero-point offset corresponded to $z=0.888$51 at $z=0.888$52, consistent with passive evolution of an old stellar population formed at $z=0.888$53. In that sense, the cluster simultaneously hosts a mature early-type galaxy population and an intracluster medium with a disturbed core, illustrating that stellar-population maturity and ICM dynamical complexity need not coincide (0910.0511).